Winning the Oil Endgame - Alternative Energy Discount House

Transcription

Winning the Oil Endgame - Alternative Energy Discount House
“We’ve embarked on the beginning of the Last Days of the Age of Oil. Nations of the world
that are striving to modernize will make choices different from the ones we have made. They
will have to. And even today’s industrial powers will shift energy use patterns . . . . [T]he market
share for carbon-rich fuels will diminish, as the demand for other forms of energy grows. And
energy companies have a choice: to embrace the future and recognize the growing demand for
a wide array of fuels; or ignore reality, and slowly—but surely—be left behind.”
— Mike Bowlin, Chairman and CEO, ARCO,
and Chairman, American Petroleum Institute,
9 Feb. 1999 1
“My personal opinion is that we are at the peak of the oil age and at the same time the beginning of the hydrogen age. Anything else is an interim solution in my view. The transition will
be very messy, and will take many and diverse competing technological paths, but the longterm future will be in hydrogen and fuel cells.”
1. Bowlin 1999.
2. Kuipers 2000.
3. Bijur, undated.
— Herman Kuipers, Business Team Manager,
Innovation & Research, Shell Global Solutions,
21 Nov. 2000 2
“The days of the traditional oil company are numbered, in part because of emerging technologies such as fuel cells. . . .”
— Peter I. Bijur, Chairman and CEO, Texaco, Inc.,
late 1990s 3
4. Ingriselli 2001.
5. Gibson-Smith 1998.
6. Fagan 2000.
“Market forces, greenery, and innovation are shaping the future of our industry and propelling
us inexorably towards hydrogen energy. Those who don’t pursue it…will rue it.”
— Frank Ingriselli, President, Texaco Technology
Ventures, 23 April 2001 4
“…[W]e’ll evolve from a world of hydrocarbon dependency to a mixture of hydrocarbon and
alternative energies use. Vast quantities of liquid hydrocarbons (oil and gas) will be left behind
in the ground, just as solid hydrocarbons (coal) are being left behind today.”
— Chris Gibson-Smith, Managing Director, BP,
25 Sept. 1998 5
“Thirty years from now there will be a huge amount of oil—and no buyers. Oil will be left in
the ground. The Stone Age came to an end, not because we had a lack of stones, and the Oil
Age will come to an end not because we have a lack of oil. . . . [Fuel cell technology] is coming
before the end of the decade and will cut gasoline consumption by almost 100 per cent. . . .On
the supply side it is easy to find oil and produce it, and on the demand side there are so many
new technologies, especially when it comes to automobiles.”
— Sheikh Zaki Yamani,
Oil Minister of Saudi Arabia (1962–86), June 2000 6
“So why is Sheikh Yamani predicting the end of the Oil Age? Because he believes that something fundamental has shifted since. . . [1973]—and, sadly for countries like Saudi Arabia, he is
quite right. Finally, advances in technology are beginning to offer a way for economies. . . to
diversify their supplies of energy and reduce their demand for petroleum, thus loosening the
grip of oil and the countries that produce it. . . .The only long-term solution . . . is to reduce the
world’s reliance on oil. Achieving this once seemed pie-in-the-sky. No longer. Hydrogen fuel
cells are at last becoming a viable alternative. . . .One day, these new energy technologies will
toss the OPEC cartel in the dustbin of history. It cannot happen soon enough.”
— “The End of the Oil Age,” editorial,
The Economist, 25 Oct. 2003
“The markets for renewable energy are the fastest growing energy markets in the world today.
***[S]uccessfully promoting renewables over the period to 2030 will prove less expensive than
. . .‘business as usual’. . . within any realistic range of real discount rates.***[T]he G8 should give
priority to efforts to trigger a step change in renewable energy markets.”
— G8 Renewable Energy Task Force, July 2001 7
“We. . . need to make great strides in transport efficiency. . . .We need to engage the consumer, not
force him or her into public transport. A European Environment minister once asked me how
to get people off their love affair with the motor car. I believe we should not even try and
interfere with that love. It is deeply imbedded, and interfering in other people’s love affairs is
seldom productive. But the love is with personal movement and space and the freedom that it
brings, not the internal combustion engine per se. We have to make eco-efficiency as fashionable
as 4-wheel-drive vehicles. We need to use the powers of social pressure and the attraction of
beautiful engineering. This is not hairshirt stuff—it should be eco-hedonism—taking pleasure
from comfort, operating performance as well as eco-efficiency.”
— Sir Mark Moody-Stuart, Chairman,
AngloAmerican, and former Chairman,
Royal Dutch/Shell Group, May 2002 8
“. . . I believe fuel cells will finally end the 100-year reign of the internal combustion engine. . . .
Fuel cells could be the predominant automotive power source in 25 years.”
— William Clay Ford, Jr., Chairman and CEO,
Ford Motor Company, 5 Oct. 2000 9
“There have already been two oil crises; we are obligated to prevent a third one. The fuel cell offers
a realistic opportunity to supplement the ‘petroleum monoculture’ over the long term. All over the
world, the auto industry is working in high gear on the fuel cell. We intend to be the market leader
in this field. Then we will have the technology, the secured patents and the jobs on our side. In
this manner, we will optimize conditions for profitable growth.”
— Jürgen Schrempp, Chairman of the Board of
Management, DaimlerChrysler, Nov. 2000 10
“General Motors absolutely sees the long-term future of the world being based on a hydrogen economy.***Forty-five percent of Fortune 50 companies will be affected, impacting almost two trillion
dollars in revenue.”
— Larry Burns, VP R&D and Planning, General Motors
Corporation, undated and 10–11 Feb. 2003 11
7. Clini & Moody-Stuart 2001,
pp. 5, 15, 9, and 7.
8. Moody-Stuart 2002.
9. Ford, Jr. 2000.
10. Berlin event with
Chancellor Schröder, quoted
in Autoweb.com.au 2000.
11. First part verified but not
dated or specifically cited
by speaker (personal communication, 25 January 2004);
second part from Burns 2003,
percentage written out.
Innovation for Profits, Jobs,
and Security
Amory B. Lovins,
E. Kyle Datta, Odd-Even Bustnes, Jonathan G. Koomey, and Nathan J. Glasgow
with
Jeff Bannon, Lena Hansen, Joshua Haacker, Jamie Fergusson, Joel Swisher PE,
Joanie Henderson, Jason Denner, James Newcomb, Ginny Hedrich, and Brett Farmery
Published by Rocky Mountain Institute
1739 Snowmass Creek Road
Snowmass, Colorado 81654-9199, USA
phone: 1.970.927.3851
fax 1.970.927.4510
www.rmi.org
This document and its detailed Technical Annex are posted at www.oilendgame.com
for free public download (for individual use).
Please send corrections and comments to [email protected].
Errata will be posted at www.oilendgame.com.
Cover art: The cover art, commissioned from Ian Naylor (www.aircrew.co.uk) on a concept by
Amory Lovins and incorporated into the cover design by RMI’s Ben Emerson,
stylizes the 13th game of the 1972 world title match between Bobby Fischer (USA) and Boris
Spassky (USSR). It shows the endgame position after 61. Be7-f8,
kindly provided by Academician R.Z. Sagdeev and reproduced at the right.
Fischer, playing Black (but shown as White in our stylized artwork),
won with a trapped rook and five passed pawns against rook, bishop, and pawn.
Acknowledgments:
This study was made possible by many people and organizations to whom the authors and publishers
are deeply grateful. Our debt to these researchers, in-kind supporters, informants,
peer-reviewers, and funders is acknowledged on pp. 266–267.
Disclaimers: This work does not express the opinion or official position of the United States Government or the U.S. Department of
Defense. Only the authors and publisher—not the sponsors, Foreword authors, or reviewers—are responsible for the views
expressed. In expressing those personal opinions and ideas, the authors and publisher are not engaged in rendering professional
services. Readers wishing personal assistance or advice should consult a competent professional. No patent liability is assumed with
respect to the use of the information contained herein. Hypercar ® and Body-in-Black™ are trademarks owned by Hypercar, Inc. Power
by the Hour™ is a trademark owned by Rolls-Royce. Scotch™ Brand Tape is a trademark of 3M. Spectra® is a trademark of Honeywell.
Chap Stick™ is a trademark owned by Wyeth Consumer Healthcare. All other trademarks, service marks, registered names, and
images are the property of their respective owners. Most product images belong to their respective manufacturers and are
downloaded from their public websites. The authors have sought to obtain permission from all image owners, and apologize to any
who could not be traced for permission and acknowledgement. Although normal scholarly care has been taken in the preparation of
this study, the authors, publisher, Foreword authors, and reviewers assume no responsibility for any errors or omissions, nor any
liability for damages resulting from the use of or reliance on information contained herein, and they specifically disclaim any
responsibility for any liability, loss, or risk, personal or otherwise, which is incurred as a consequence, directly or indirectly, of the use
or application of any of the contents of this work. We accept no responsibility or liability for the contents of pages accessible through
links to our webpages. This report is unclassified and contains no confidential or proprietary data.
The publisher and authors The publisher and authors are described on pp. 282–285.
The senior author’s and publisher’s interest is declared on pages 61 and 63.
Units of measure are customary U.S. units with some metric supplements; see also p. 39.
Production Credits
Editor: Beatrice Aranow
Art Director, Graphic Designer: Ben Emerson
Production Assistants: Jenny Constable, Vinay Gupta, Anne Jakle, Tyler Lindsay, Morley McBride, Ann Mason, Christina Page
Webmaster (www.oilendgame.com): William Simon
Type: Centaur (primary), Palatino (main text) and Univers (supporting text elements)
Paper: Domtar Inc. Cornwall ® 80#, 50% recycled, 15% PCW, PCF (cover) and New Leaf ™ Eco Offset 60#, 100% PCW, PCF (text)
Printed with vegetable-based ink in the United States of America
© 2004 Rocky Mountain Institute. All rights reserved.
The financial sponsors of this study may freely reproduce and distribute this report for noncommercial purposes
with acknowledgement to Rocky Mountain Institute.
First Edition
ISBN 1-881071-10-3
Winning the Oil Endgame
O ve r v i e w C o n te n t s
Detailed Contents . . . . . . . . . . . . . . . . . . iv
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . vii
Executive Summary . . . . . . . . . . . . . . . . ix
Forewords . . . . . . . . . . . . . . . . . . . . . . . xv
George P. Shultz
xv
Sir Mark Moody-Stuart
xviii
www.oilendgame.com
Frontispiece . . . . . . . . . . . . . . . . xxiii–xxiv
Oil Dependence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
This Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Saving Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Substituting for Oil . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Combined Conventional Potential . . . . . . . . . . . . . 123
Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Implications and Conclusions . . . . . . . . . . . . . . . . . 243
Image credits . . . . . . . . . . . . . . . . . . . . 276
Table of figures and tables
. . . . . . . . . 277
Acknowledgments . . . . . . . . . . . . . . . . 278
About the publisher and authors . . . . . 282
References . . . . . . . . . . . . . . . . . . . . . . 286
Other RMI publications
307
Detailed Contents
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Boxes
Oil is the lifeblood of modern industrial economies—but not forever 1
Even an important industry can be displaced by competitors 4
1: An example of domestic
energy vulnerability (p. 12)
2: Oil is fungible (p. 14)
3: The uncounted economic
cost of oil-price volatility
(p. 16)
4: Hedging the risk of oil
depletion (p. 24)
America can replace oil quickly—and already has 7
Oil supplies are becoming more concentrated and less secure 8
Domestic oil is limited 12
Counting the direct cost of oil dependence 15
Oil dependence’s hidden costs may well exceed its direct costs 17
Petrodollars tend to destabilize 18
Sociopolitical instability drives military costs 19
Nonmilitary societal costs 21
Adding up the hidden costs 22
Could less oil dependence be not only worthy but also profitable? 26
Beliefs that hold us back 26
Whatever exists is possible 29
This Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5: Conventions (p. 39)
Structure and methodology 33
Conservatisms and conventions 37
Saving Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Option 1. Efficient use of oil 43
6: How do light vehicles
use fuel, and how can they
save fuel? (p. 46)
7: Superefficient but
uncompromised (p. 62)
8: Analyzing an ultralight
hybrid’s efficiency (p. 68)
9: Analyzing and extending
ultralight vehicle costs
(p. 69)
10: Comparing light-vehicle
prices (p. 72)
11: Saving oil in existing
military platforms (p. 86)
Transportation 44
Light vehicles 44
The conventional view 49
Advanced automotive technologies: lightweight, low-drag, highly integrated 52
Drag and rolling resistance 52
Lightweighting: the emerging revolution 53
Ultralight but ultrastrong 57
Applying ultralight materials 61
Lighter-but-safer vehicles dramatically extend cheap oil savings 64
Heavy trucks 73
Medium trucks 77
Intelligent highway systems (IHS) 78
Other civilian highway and off-road vehicles 79
Trains 79
Ships 79
Airplanes 79
Military vehicles 84
The fuel logistics burden 84
Military efficiency potential 85
Feedstocks and other nonfuel uses of oil 93
Industrial fuel 97
Buildings 97
Electricity generation 98
Combined efficiency potential 99
iv
D e t a i l e d C o n t e n t s (continued)
Substituting for Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Option 2. Substituting biofuels and biomaterials 103
The input side: biomass feedstocks and rural economies 107
Biomaterials 110
Boxes (continued)
Option 3. Substituting saved natural gas 111
Overview 111
Saving natural gas 112
Electric utilities 113
Buildings 115
Industrial fuel 115
Substituting saved gas for oil 117
12: Replacing one-third of
remaining non-transportation
oil use with saved natural gas
(p. 118)
Combined Conventional Potential . . . . . . . . . . . . . . . . . 123
Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Strategic vision 127
The prize 127
Vaulting the barriers 128
The endangered automotive sector—why it’s important to act now 130
Four competitive threats 132
China and India 135
Suppressing the signals 136
Crafting an effective energy strategy: transformative business innovation 137
The creative destruction dilemma 138
Business challenges: market, business, and customer realities 139
Most U.S. light-vehicle buyers scarcely value fuel economy 139
Most firms underinvest in energy efficiency too 141
The risk of being risk-averse 143
Business opportunities: competitive strategy for profitable transformation in the
transportation sector 145
Lowering light vehicles’ manufacturing risk 146
Manufacturing investment and variable cost 146
Market adoption 149
Restoring profitability in the trucking sector 150
Revitalizing the airline and airplane industries 154
Getting generation-after-next planes off the ground 157
Creating a new high-technology industrial cluster 159
Restoring farming, ranching, and forest economies 162
13: Guilt-free driving:
hybrid cars enter the market
(p. 131)
14: Opening moves: Boeing’s
bet on fuel efficiency as the
future for commercial aircraft
(p. 133)
15: Radically simpler
automaking with advanced
composites (p. 147)
16: Flying high:
fuel savings arbitrage
(p. 156)
If we don’t act soon, the invisible hand will become the invisible fist 166
(Implementation continued next page)
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
v
D e t a i l e d C o n t e n t s (continued)
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
(Implementation continued from previous page)
Crafting coherent supportive policies 169
Boxes (continued)
17: Gridlock as Usual
according to Thucycides,
ca. 431–404 BCE (p. 170)
18: Modeling the effects
of policy on light vehicle
sales and stocks (p. 182)
19: How feebates work
(p. 186)
20: More antidotes to
regressivity (p. 196)
21: Golden Carrots: theme
and variations (p. 200)
22: Realigning auto-safety
policy with modern
engineering (p. 207)
23: Pay-at-the-Pump
car insurance (p. 218)
Government’s role in implementation 169
Fuel taxes 173
Standards, mandates, and quotas 175
Federal policy recommendations for light vehicles 178
Feebates 186
Low-income scrap-and-replace program 191
Smart government fleet procurement 197
“Golden Carrots” and technology procurement 199
“Platinum Carrot” advanced-technology contest 201
Supporting automotive retooling and retraining 203
R&D and early military procurement 204
Automotive efficiency and safety regulation: first, do no harm 206
Other federal policy recommendations 208
Supporting investment in domestic energy supply infrastructure 208
Heavy-vehicle policy 211
Aircraft policy 212
Other transportation policy 212
Shifting taxation from fuel to roads and driving 212
Integrating transportation systems 214
Is this trip necessary (and desired)? 214
Non-transportation federal policy 215
States: incubators and accelerators 216
Transportation 216
Electricity and natural-gas pricing 219
Renewable energy 220
Military policy: fuel efficiency for mission effectiveness 221
Civil preparedness: evolving toward resilience 222
Civil society: the sum of all choices 223
Beyond gridlock: changing politics 225
Option 4. Substituting hydrogen 227
Beyond mobilization to a basic shift in primary energy supply 227
Hydrogen: practical after all 230
Eight basic questions 233
Why is hydrogen important, and is it safe? 233
How would a light vehicle safely and affordably store enough hydrogen
to drive 300+ miles? 233
Under what conditions is hydrogen a cheaper light-vehicle fuel than oil? 234
What’s the cheapest way to produce and deliver hydrogen to meet the economic
conditions required for adoption? 236
Are there enough primary energy sources for this transition? 238
What technologies are required to enable the hydrogen transition? 241
How can the U.S. profitably make the transition from oil to hydrogen? 241
When could this transition occur? 242
vi
D e t a i l e d C o n t e n t s (continued)
Implications and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 243
Implications 243
Employment 243
Allies and trading partners 244
Developing countries 245
Leapfrog development 245
Climate change and development 246
The global economy, oil savings, and development 247
Oil-exporting countries 248
The creative destruction challenge for oil companies 250
Other energy industries 257
Boxes (continued)
24: Shell’s visionary energy
futures (p. 252)
25: What about nuclear power?
(p. 258)
U.S. military force structure, posture, and doctrine 261
Toward a new strategic doctrine 262
U.S. federal budget 265
Environment, public health, and quality of life 268
Conclusions 271
Abstract:
This independent, peer-reviewed synthesis for American business and military leaders charts a roadmap for getting the
United States completely, attractively, and profitably off oil. Our strategy integrates four technological ways to displace oil:
using oil twice as efficiently, then substituting biofuels, saved natural gas, and, optionally, hydrogen. Fully applying today’s
best efficiency technologies in a doubled-GDP 2025 economy would save half the projected U.S. oil use at half its forecast
cost per barrel. Non-oil substitutes for the remaining consumption would also cost less than oil. These comparisons conservatively assign zero value to avoiding oil’s many “externalized” costs, including the costs incurred by military insecurity,
rivalry with developing countries, pollution, and depletion. The vehicle improvements and other savings required needn’t be
as fast as those achieved after the 1979 oil shock.
The route we suggest for the transition beyond oil will expand customer choice and wealth, and will be led by business for
profit. We propose novel public policies to accelerate this transition that are market-oriented without taxes and innovationdriven without mandates. A $180-billion investment over the next decade will yield $130-billion annual savings by 2025;
revitalize the automotive, truck, aviation, and hydrocarbon industries; create a million jobs in both industrial and rural
areas; rebalance trade; make the United States more secure, prosperous, equitable, and environmentally healthy;
encourage other countries to get off oil too; and make the world more developed, fair, and peaceful.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
vii
Executive Summary
W
inning the Oil Endgame offers a coherent strategy for ending oil
dependence, starting with the United States but applicable worldwide. There are many analyses of the oil problem. This synthesis
is the first roadmap of the oil solution—one led by business for profit,
not dictated by government for reasons of ideology. This roadmap is independent, peer-reviewed, written for business and military leaders, and
co-funded by the Pentagon. It combines innovative technologies and new
business models with uncommon public policies: market-oriented without
taxes, innovation-driven without mandates, not dependent on major (if any)
national legislation, and designed to support, not distort, business logic.
Two centuries ago, the first industrial revolution made people a hundred
times more productive, harnessed fossil energy for transport and production, and nurtured the young U.S. economy. Then, over the past 145 years,
the Age of Oil brought unprecedented mobility, globe-spanning military
power, and amazing synthetic products.
But at what cost? Oil, which created the sinews of our strength, is now
becoming an even greater source of weakness: its volatile price erodes
prosperity; its vulnerabilities undermine security; its emissions destabilize
climate. Moreover the quest to attain oil creates dangerous new rivalries
and tarnishes America’s moral standing. All these costs are rising. And
their root causes—most of all, inefficient light trucks and cars—also threaten the competitiveness of U.S. automaking and other key industrial sectors.
The cornerstone of the next industrial revolution is therefore winning the
Oil Endgame. And surprisingly, it will cost less to displace all of the oil
that the United States now uses than it will cost to buy that oil. Oil’s current market price leaves out its true costs to the economy, national security, and the environment. But even without including these now “externalized” costs, it would still be profitable to displace oil completely over the
next few decades. In fact, by 2025, the annual economic benefit of that displacement would be $130 billion gross (or $70 billion net of the displacement’s costs). To achieve this does not require a revolution, but merely
consolidating and accelerating trends already in place: the amount of oil
the economy uses for each dollar of GDP produced, and the fuel efficiency of light vehicles, would need only to improve about three-fifths as
quickly as they did in response to previous oil shocks.
Saving half the oil America uses, and substituting cheaper alternatives
for the other half, requires four integrated steps:
• Double the efficiency of using oil. The U.S. today wrings twice as
much work from each barrel of oil as it did in 1975; with the latest
proven efficiency technologies, it can double oil efficiency all over
again. The investments needed to save each barrel of oil will cost
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
ix
Executive Summary
Four integrated steps • Double the efficiency of using oil (continued)
only $12 (in 2000 $), less than half the officially forecast $26 price of
that barrel in the world oil market. The most important enabling
technology is ultralight vehicle design. Advanced composite or
lightweight-steel materials can nearly double the efficiency of
today’s popular hybrid-electric cars and light trucks while improving safety and performance. The vehicle’s total extra cost is repaid
from fuel savings in about three years; the ultralighting is approximately free. Through emerging manufacturing techniques, such
vehicles are becoming practical and profitable; the factories to produce them will also be cheaper and smaller.
• Apply creative business models and public policies to speed the
profitable adoption of superefficent light vehicles, heavy trucks,
and airplanes. Combined with more efficient buildings and factories, these efficient vehicles can cut the official forecast of oil use by
29% in 2025 and another 23% soon thereafter—52% in all. Enabled
by a new industrial cluster focusing on lightweight materials, such
as carbon-fiber composites, such advanced-technology vehicles can
revitalize these three strategic sectors and create important new
industries.
• Provide another one-fourth of U.S. oil needs by a major domestic
biofuels industry. Recent advances in biotechnology and celluloseto-ethanol conversion can double previous techniques’ yield, yet
cost less in both capital and energy. Replacing fossil-fuel hydrocarbons with plant-derived carbohydrates will strengthen rural America, boost net farm income by tens of billions of dollars a year, and
create more than 750,000 new jobs. Convergence between the energy,
chemical, and agricultural value chains will also let versatile new
classes of biomaterials replace petrochemicals.
• Use well established, highly profitable efficiency techniques to
save half the projected 2025 use of natural gas, making it again
abundant and affordable, then substitute part of the saved gas for
oil. If desired, the leftover saved natural gas could be used even
more profitably and effectively by converting it to hydrogen,
displacing most of the remaining oil use—and all of the oil use
if modestly augmented by competitive renewable energy.
These four shifts are fundamentally disruptive to current business models.
They are what economist Joseph Schumpeter called “creative destruction,”
where innovations destroy obsolete technologies, only to be overthrown in
turn by ever newer, more efficient rivals. In The Innovator’s Dilemma, Harvard
Business School professor Clayton Christensen explained why industry
leaders often get blindsided by disruptive innovations—technological
gamechangers—because they focus too much on today’s most profitable
customers and businesses, ignoring the needs of the future. Firms that are
x
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Executive Summary
quick to adopt innovative technologies and business models will be the
winners of the 21st century; those that deny and resist change will join
the dead from the last millennium. In the 108-year history of the Dow Jones
Industrial Average, only one of 12 original companies remains a corporate
entity today—General Electric. The others perished or became fodder
for their competitors.
What policies are needed? American companies can be among the quick
leaders in the 21st century, but it will take a cohesive strategy-based
transformation, bold business and military leadership, and supportive
government policies at a federal or at least a state level. Winning the Oil
Endgame charts these practical steppingstones to an oil-free America:
• Most importantly, revenue- and size-neutral “feebates” can shift
customer choice by combining fees on inefficient vehicles with
rebates to efficient vehicles. The feebates apply separately within
each vehicle-size class, so freedom of choice is unaffected. Indeed,
choice is enhanced as customers start to count fuel savings over
the vehicle’s life, not just the first few years, and this new pattern
of demand pulls superefficient but uncompromised vehicles from
the drawing-board into the showroom.
• A scrap-and-replace program can lease or sell super-efficient cars
to low-income Americans—on terms and with fuel bills they can
afford—while scrapping clunkers. This makes personal mobility
affordable to all, creates a new million-car-a-year market for the
new efficiency technologies, and helps clean our cities’ air.
• Military needs for agility, rapid deployment, and streamlined
logistics can drive Pentagon leadership in developing key
technologies.
• Implementing smart government procurement and targeted technology acquisition (the “Golden Carrot”) for aggregated buyers will
accelerate manufacturers’ conversion, while a government-sponsored $1-billion prize for success in the marketplace, the “Platinum
Carrot,” will speed development of even more advanced vehicles.
• To support U.S. automakers’ and suppliers’ need to invest about
$70 billion to make advanced technology vehicles, federal loan guarantees can help finance initial retooling where needed; the investments should earn a handsome return, with big spin-off benefits.
• Similar but simpler policies—loan guarantees for buying efficient
new airplanes (while scrapping inefficient parked ones), and better
information for heavy truck buyers to spur market demand for
doubled-efficiency trucks—can speed these oil-saving innovations
from concept to market.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
xi
Executive Summary
Practical steppingstones to an oil-free America (continued)
• Other policies can hasten competitive evolution of next-generation
biofuels and biomaterials industries, substituting durable revenues
for dwindling agricultural subsidies, and encouraging practices
that protect both topsoil and climate.
What happens to the oil industry? The transition beyond oil is already starting to transform oil companies like Shell and BP into energy companies.
Done right, this shift can profitably redeploy their skills and assets rather
than lose market share. Biofuels are already becoming a new product line
that leverages existing retail and distribution infrastructure and can attract
another $90 billion in biofuels and biorefining investments. By following
this roadmap, the U.S. would set the stage by 2025 for the checkmate move
in the Oil Endgame—the optional but advantageous transition to a hydrogen economy and the complete and permanent displacement of oil as a
direct fuel. Oil may, however, retain or even gain value as one of the competing sources of hydrogen.
How big is the prize? Investing $180 billion over the next decade to eliminate
oil dependence and revitalize strategic industries can save $130 billion gross,
or $70 billion net, every year by 2025. This saving, equivalent to a large tax
cut, can replace today’s $10-billion-a-month oil imports with reinvestments
in ourselves: $40 billion would pay farmers for biofuels, while the rest
could return to our communities, businesses, and children. Several million
automotive and other transportation-equipment jobs now at risk can be
saved, and one million net new jobs can be added across all sectors. U.S.
automotive, trucking, and aircraft production can again lead the world,
underpinned by 21st century advanced-materials and fuel-cell industries.
A more efficient and deployable military could refocus on its core mission—
protecting American citizens rather than foreign supply lines—while supporting and deploying the innovations that eliminate oil as a cause of conflict. Carbon dioxide emissions will shrink by one-fourth with no additional
cost or effort. The rich-poor divide can be drastically narrowed at home by
increased access to affordable personal mobility, shrinking the welfare rolls,
and abroad by leapfrogging over oil-dependent development patterns.
The U.S. could treat oil-rich countries the same as countries with no oil.
Being no longer suspected of seeking oil in all that it does in the world
would help to restore U.S. moral leadership and clarity of purpose.
While the $180-billion investment needed is significant, the United States’
economy already pays that much, with zero return, every time the oil
price spikes up as it has done in 2004. (And that money goes into OPEC’s
coffers instead of building infrastructure at home.) Just by 2015, the early
steps in this proposed transition will have saved as much oil as the U.S.
gets from the Persian Gulf. By 2040, oil imports could be gone. By 2050,
the U.S. economy should be flourishing with no oil at all.
xii
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Executive Summary
How do we get started? Every sector of society can contribute to this national project. Astute business leaders will align their corporate strategies and
reorganize their firms and processes to turn innovation from a threat to
a friend. Military leaders will speed military transformation by promptly
laying its foundation in superefficient platforms and lean logistics. Political
leaders will craft policies that stimulate demand for efficient vehicles,
reduce R&D and manufacturing investment risks, support the creation of
secure domestic fuel supplies, and eliminate perverse subsidies and regulatory obstacles. Lastly, we, the people, must play a role—a big role—
because our individual choices guide the markets, enforce accountability,
and create social innovation.
Our energy future is choice, not fate. Oil dependence is a problem we need
no longer have—and it’s cheaper not to. U.S. oil dependence can be eliminated by proven and attractive technologies that create wealth, enhance
choice, and strengthen common security. This could be achieved only
about as far in the future as the 1973 Arab oil embargo is in the past. When
the U.S. last paid attention to oil, in 1977–85, it cut its oil use 17% while
GDP grew 27%. Oil imports fell 50%, and imports from the Persian Gulf
by 87%, in just eight years. That exercise of dominant market power—
from the demand side—broke OPEC’s ability to set world oil prices for a
decade. Today we can rerun that play, only better. The obstacles are less
important than the opportunities if we replace ignorance with insight,
inattention with foresight, and inaction with mobilization. American business can lead the nation and the world into the post-petroleum era, a
vibrant economy, and lasting security—if we just realize that we are the
people we have been waiting for.
Together we can end oil dependence forever.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
xiii
Foreword
C
rude prices are rising, uncertainty about developments in the
Middle East roils markets and, well, as Ronald Reagan might say,
“Here we go again.”
Once more we face the vulnerability of our oil supply to political disturbances. Three times in the past thirty years (1973, 1978, and 1990) oil price
spikes caused by Middle East crises helped throw the U.S. economy into
recession. Coincident disruption in Venezuela and Russia adds to unease,
let alone prices, in 2004. And the surging economies of China and India are
contributing significantly to demand. But the problem far transcends economics and involves our national security. How many more times must we
be hit on the head by a two-by-four before we do something decisive about
this acute problem?
In 1969, when I was Secretary of Labor, President Nixon made me the chairman of a cabinet task force to examine the oil import quota system, in place
since 1954. Back then, President Eisenhower considered too much dependence on imported oil to be a threat to national security. He thought anything
over 20 percent was a real problem. No doubt he was nudged by his friends
in the Texas and Louisiana oil patches, but Ike was no stranger to issues of
national security and foreign policy.
The task force was not prescient or unanimous but, smelling trouble, the
majority could see that imports would rise and they recommended a new
monitoring system to keep track of the many uncertainties we could see
ahead, and a new system for regulating imports. Advocates for even greater
restrictions on imports argued that low-cost oil from the Middle East would
flood our market if not restricted.
By now, the quota argument has been stood on its head as imports make
up an increasing majority, now almost 60 percent and heading higher, of
the oil we consume. And we worry not about issues of letting imports in
but that they might be cut off. Nevertheless, the point about the importance
of relative cost is as pertinent today as back then and applies to the competitive pressures on any alternative to oil. And the low-cost producers of oil
are almost all in the Middle East.
A native of New York,
George P. Shultz graduated
from Princeton University
in 1942. After serving in
the Marine Corps (1942–45),
he earned a PhD at MIT.
Mr. Shultz taught at MIT
and the University of
Chicago Graduate School
of Business, where he
became dean in 1962.
He was appointed Secretary of Labor in 1969,
Director of the Office of
Management and Budget in
1970, and Secretary of the
Treasury in 1972. From 1974
to 1982, he was President
of Bechtel Group, Inc. Mr.
Shultz served as Chairman
of the President’s Economic
Policy Advisory Board
(1981–82) and Secretary of
State (1982–89). He is chairman of the J.P. Morgan
Chase International Council
and the Accenture Energy
Advisory Board. Since 1989,
he has been a Distinguished
Fellow at the Hoover Institution, Stanford University.
That is an area where the population is exploding out of control, where
youth is by far the largest group, and where these young people have little
or nothing to do. The reason is that governance in these areas has failed
them. In many countries, oil has produced wealth without the effort that
connects people to reality, a problem reinforced in some of them by the fact
that the hard physical work is often done by imported labor. The submissive
role forced on women has led to this population explosion. A disproportionate share of the world’s many violent conflicts is in this area. So the Middle
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
xv
Foreword (continued) George P. Shultz
East remains one of the most unstable parts of the world. Only a dedicated
optimist could believe that this assessment will change sharply in the near
future. What would be the impact on the world economy of terrorist sabotage of key elements of the Saudi pipeline infrastructure?
I believe that, three decades after the Nixon task force effort, it is long past
time to take serious steps to alter this picture dramatically. Yes, important
progress has been made, with each administration announcing initiatives to
move us away from oil. Advances in technology and switches from oil to
natural gas and coal have caused our oil use per dollar of GDP to fall in half
since 1973. That helps reduce the potential damage from supply problems.
But potential damage is increased by the rise of imports from 28 percent to
almost 60 percent of all the oil we use. The big growth sector is transportation, up by 50 percent. Present trends are unfavorable; if continued, they
mean that we are likely to consume—and import—several million barrels a
day more by 2010.
Beyond U.S. consumption, supply and demand in the world’s oil market
has become tight again, leading to many new possibilities of soaring oil
prices and massive macroeconomic losses from oil disruptions. We also
have environmental problems to concern us. And, most significantly, our
national security and its supporting diplomacy are left vulnerable to fears
of major disruptions in the market for oil, let alone the reality of sharp price
spikes. These costs are not reflected in the market price of oil, but they are
substantial.
What more can we do? Lots, if we are ready for a real effort. I remember
when, as Secretary of the Treasury, I reviewed proposals for alternatives to
oil from the time of the first big oil crisis in 1973. Pie in the sky, I thought.
But now the situation is different. We can, as Amory Lovins and his colleagues show vividly, win the oil endgame.
How do we go about this? A baseball analogy may be applicable. Fans
often have the image in their minds of a big hitter coming up with the
bases loaded, two outs, and the home team three runs behind. The big hitter wins the game with a home run. We are addicted to home runs, but the
outcome of a baseball game is usually determined by a combination of
walks, stolen bases, errors, hit batsmen, and, yes, some doubles, triples, and
home runs. There’s also good pitching and solid fielding, so ball games are
won by a wide array of events, each contributing to the result. Lovins and
his coauthors show us that the same approach can work in winning the oil
endgame. There are some potential big hits here, but the big point is that
there are a great variety of measures that can be taken that each will contribute to the end result. The point is to muster the will power and drive to
pursue these possibilities.
xvi
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Foreword (continued) George P. Shultz
How do we bring that about? Let’s not wait for a catastrophe to do the job.
Competitive information is key. Our marketplace is finely tuned to the
desire of the consumer to have real choices. We live in a real information
age, so producers have to be ready for the competition that can come out of
nowhere. Lovins and his colleagues provide a huge amount of information
about potential competitive approaches. There are home run balls here, the
ultimate one being the hydrogen economy. But we don’t have to wait for
the arrival of that day. There are many things that can be done now, and this
book is full of them. Hybrid technology is on the road and currently increases gas mileage by 50 percent or more. The technology is scaleable. This
report suggests many ways to reduce weight and drag, thereby improving
performance. A big point in this report is evidence that new, ultralight-butsafe materials can nearly redouble fuel economy at little or no extra cost.
Sequestration of effluent from use of coal may be possible on an economic
and comfortable basis, making coal a potentially benign source of hydrogen.
Maybe hydrogen could be economically split out of water by electrolysis,
perhaps using renewables such as windpower; or it could certainly be
made, as nearly all of it is now, by natural gas saved from currently wasteful
practices, an intriguingly lucrative option often overlooked in discussions of
today’s gas shortages. An economy with a major hydrogen component
would do wonders for both our security and our environment. With evident
improvements in fuel cells, that combination could amount to a very big
deal. Applications include stationary as well as mobile possibilities, and
other ideas are in the air. Real progress has been made in the use of solar
systems for heat and electricity. Scientists, technologists, and commercial
organizations in many countries are hard at work on these issues.
Sometimes the best way to get points across is to be provocative, to be a bull
in a china closet. Amory Lovins loves to be a bull in a china closet—anybody’s china closet. With this book, the china closet he’s bursting into is ours
and we should welcome him because he is showing us how to put the closet
back together again in far more satisfactory form. In fact, Lovins and his
team make an intriguing case that is important enough to merit careful attention by all of us, private citizens and business and political leaders alike.
— George P. Shultz
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
xvii
Foreword
I
Born in Antigua, Mark
Moody-Stuart earned a
doctorate in geology in 1966
at Cambridge, then worked
for Shell starting as an
exploration geologist, living
in the Netherlands, Spain,
Oman, Brunei, Australia,
Nigeria, Turkey, Malaysia,
and the UK, and retiring as
Chairman of the Royal
Dutch/Shell Group in 2001.
He is Chairman of Anglo
American plc, a Director
of HSBC and of Accenture,
a Governor of Nuffield
Hospitals, President of the
Liverpool School of Tropical
Medicine, and on the
board of the Global Reporting Initiative and the
International Institute for
Sustainable Development.
He is Chairman of the
Global Business Coalition
for HIV/AIDS and Co-Chair
of the Singapore British
Business Council. He was
Co-Chair of the G8 Task
Force on Renewable
Energy (2000–01) and
Chairman of Business
Action for Sustainable
Development, an initiative
of the ICC and the World
Business Council for
Sustainable Development
before and during the
2002 World Summit on
Sustainable Development
in Johannesburg. During
2001–04 he served on the
UN Secretary General’s
Advisory Council for the
Global Compact. He was
knighted in 2000. With his
wife Judy, he drives a
Toyota Prius and is an
investor in Hypercar, Inc.
xviii
n this compelling synthesis, Amory Lovins and his colleagues at Rocky
Mountain Institute provide a clear and penetrating view of one of the
critical challenges facing the world today: the use of energy, especially
oil, in transportation, industry, buildings, and the military. This report
demonstrates that innovative technologies can achieve spectacular savings in all of these areas with no loss of utility, convenience and function.
It makes the business case for how a profitable transition for the automotive, truck, aviation, and oil sectors can be achieved, and why they
should embrace technological innovation rather than be destroyed by it.
We are not short of energy in this world of ours; we have large resources
of the convenient hydrocarbons on which our economies are based, and
even greater resources of the coal on which our economies were originally
built. But there are two serious issues relating to its supply and use.
First, some three fourths of the reserves sit in a few countries of the
Middle East, subject to actual and potential political turmoil. Second,
there are the long-term climatic effects of the burning of increasing
amounts of fossil fuels. While the normal rate of change of technology
is likely to mean that we will be on one of the lower impact scenarios of
climate change and not at the apocalyptic end favoured by doom mongers, it is reasonably certain that our world will have to adapt to significant climate change over the next century. These two factors mean that,
unless there is a change of approach, the United States will inexorably
become increasingly dependent on imported energy—be it oil or natural
gas. At the same time, on the international scene, the United States will
be criticised by the rest of the world for profligate use of energy, albeit to
fuel an economy on whose dynamism and success the rest of the world
is also manifestly dependent. Furthermore, thoughtful people wonder
what we will do if the booming economies and creative people of China
and India have energy demands which are on the same development
curve as the United States.
The RMI team has approached this economic and strategic dilemma with
technical rigour, good humour, and common sense, while addressing two
key requirements often overlooked by energy policy advocates.
First, we have to deliver the utility, reliability and convenience that the
consumer has come to expect. As business people we recognise this. It is
no good expecting people in the United States to suddenly drive smaller,
less convenient or less safe vehicles. We have to supply the same comfort
and utility at radically increased levels of energy efficiency. Most consumers, who are also voters, have only a limited philosophical interest in
energy efficiency, security of supply, and climate change. Most of us have a
very intense interest in personal convenience and safety—we expect gov-
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Foreword (continued) Sir Mark Moody-Stuart
ernments and business to handle those other issues on our behalf. There is
a very small market in this world for hair shirts. Similarly, we cannot
expect the citizens of China and India to continue to ride bicycles in the
interests of the global environment. They have exactly the same aspirations to comfort and convenience as we do. This book demonstrates how
by applying existing technologies to lightweight vehicles with the use of
composites, by the use of hybrid powertrains already in production, and
with the rapid evolution to new technologies, we can deliver the high levels of convenience and reliability we are used to at radically increased levels of energy efficiency, while also maintaining cost efficiency.
The second critical requirement is that the process of transition should be
fundamentally economic. We know in business that while one may be
prepared to make limited pathfinding investments at nil or low return in
order to develop new products and markets, this can not be done at a
larger scale, nor indefinitely. What we can do, and have seen done repeatedly, is to transform markets by delivering greater utility at the same cost
or the same utility at a lower cost, often by combining more advanced
technologies with better business models. When this happens, the rate of
change of markets normally exceeds our wildest forecasts and within a
space of a few years a whole new technology has evolved.
A good example of the rapid development and waning of technology is
the fax machine. With astonishing rapidity, because of its functional
advantages over surface mail, the fax machine became globally ubiquitous.
The smallest businesses around the world had one and so did numerous
homes. The fax has now become almost obsolete because of e-mail, the email attachment and finally the scanned e-mail attachment. The connectivity of the Internet, of which e-mail is an example, has transformed the way
we do business. What this book shows is that the delivery of radically
more energy-efficient technologies has dramatic cost implications and
therefore has the potential for a similarly economically driven transition.
The refreshingly creative government policies suggested here to smooth
and speed that transition are a welcome departure from traditional
approaches that often overlook or even reject the scope of enterprise to be
an important part of the solution. These innovative policies, too, merit
serious attention, especially as an integrated package, and I suspect they
could win support across the political spectrum.
The technological, let alone policy, revolution has not been quick in coming to the United States. Yet as has happened before in the automobile
industry, others are aware of the potential of the technology. Perhaps
because of Japan’s obsession with energy security, Toyota and Honda
began some years ago to hone the electric-hybrid technology that is likely
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
xix
Foreword (continued) Sir Mark Moody-Stuart
to be an important part of the energy efficiency revolution. As a result,
U.S. automobile manufacturers who now see the market opportunities of
these technologies are turning to the proven Japanese technology to deliver it rapidly.
I believe that we may see a similar leapfrogging of technology from
China. China is fully aware of the consequences on energy demand, energy imports, and security of supply of its impressive economic growth.
Already China is using regulation to channel development into more
energy-efficient forms. The burgeoning Chinese automobile industry is
likely to be guided down this route—delivering the function and convenience, but at greatly increased levels of efficiency. And it is not just in the
automobile industry—by clearly stated national policy it applies to all
areas of industrial activity. This has great implications both for the participation by U.S. firms in investment in China, and also in the impact of
future Chinese manufactures on a global market that is likely to be paying much greater attention to energy efficiency.
As a businessman, I am attracted by the commercial logic and keen
insight that this report brings to the marketplace struggle between oil and
its formidable competitors on both the demand and the supply sides.
Indeed, during my time in both Shell and AngloAmerican, RMI’s engineers have helped ours to confirm unexpectedly rich deposits of mineable
“negawatts” and “negabarrels” in our own operations—an exploration
effort we’re keen to intensify to the benefit of both our shareholders and
the environment.
As a lifelong oil man and exploration geologist, I am especially excited
to learn about the Saudi Arabia-size riches that Amory Lovins and RMI’s
explorers have discovered in what they term the Detroit Formation—
through breakthrough vehicle design that can save vast amounts of oil
more cheaply than it can be supplied. And as a citizen and grandparent,
I am pleased that RMI proposes new business models to span the mobility divide that separates rich and poor, not just in the United States,
but in many places in the world. Concern about such opportunity divides
is increasingly at the core not just of international morality but also of
stability and peace.
xx
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Foreword (continued) Sir Mark Moody-Stuart
This book points the way to an economically driven energy transformation. And its subtitle “Innovation for Profit, Jobs, and Security” is both a
prospectus for positive change and a reminder that both the United States
and other countries can be rapid adapters of innovative technologies,
with equally transformative economic consequences. As someone who
has spent a lifetime involved in energy and changes in energy patterns,
I find the choice an easy one to make. The global economy is very much
dependent on the health of the U.S. economy, so I hope that the U.S.
indeed makes the right choice.
This report will help to launch, inspire, and inform a new and necessary
conversation about energy and security, economy and environment.
Its outcome is vital for us all.
— Sir Mark Moody-Stuart
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
xxi
Innovation for Profits, Jobs,
and Security
Winning the Oil Endgame:
Innovation for Profits, Jobs, and Security
“On our present course, America 20 years from now will import nearly two of every three
barrels of oil—a condition of increased dependency on foreign powers that do not always
have America’s interests at heart.***But it is not beyond our power to correct. America
leads the world in scientific achievement, technical skill, and entrepreneurial drive. Within
our country are abundant natural resources, unrivaled technology, and unlimited human
creativity. With forward-looking leadership and sensible policies, we can meet our future
energy demands and promote energy [efficiency]…and do so in environmentally responsible ways that set a standard for the world.***Per capita oil consumption, which reached a
peak in 1978, has fallen 20 percent from that level.***Today’s automobiles…use about 60
percent of the gasoline [per mile] they did in 1972, while new refrigerators require just
one-third the electricity they did 30 years ago. As a result, since 1973, the U.S. economy
has grown by 126 percent, while energy use has increased by only 30 percent. In the
1990s alone, manufacturing output expanded by 41 percent, while industrial electricity
consumption grew by only 11 percent. We must build on this progress and strengthen
America’s commitment to energy efficiency***[which] helps the United States reduce energy imports, the likelihood of energy shortages, emissions, and the volatility of energy
prices.***Our country has met many great tests. Some have imposed extreme hardship and
sacrifice. Others have demanded only resolve, ingenuity, and clarity of purpose. Such is
the case with energy today.”
— National Energy Policy Development Group;
Reliable, Affordable, and Environmentally Sound
Energy for America’s Future; 200112
“If you want to build a ship, don’t drum up the men to gather wood, divide the work and give orders.
Instead, teach them to yearn for the vast and endless sea.”
“As for the future, your task is not to foresee it, but to enable it.”
— Antoine de Saint-Exupéry 13
12. National Energy Policy Development Group 2001,
pp. x, viii–ix, 1–10, xi–xii, 1–4, and xv.
13. de Saint-Exupéry 1948.
Oil Dependence
H
ow can America—all Americans together—win the oil endgame?
Is a world beyond oil imaginable? Practical? Profitable?
Why might it be prudent to get there sooner rather than later?
How could we get there if we wanted to?
How could this build a stronger country and a safer world?
At the start of World War II, Detroit switched in six months from making
four million cars a year to making the tanks and aircraft that won the
war.14 Today, even absent the urgency of those dark days, American
industry could advantageously launch a transition to different cars and
other tools for making oil use stabilize, dwindle, and in a few generations become but a memory in museums. Some farsighted oil and car
companies already envisage such a future (see inside front cover). Some
are striving to create it before their rivals discover how major asset redeployments can yield more profit and less risk. These business leaders
see that if government steers, not rows, then competitive enterprise, supported by judicious policy and vibrant civil society, can turn the insoluble oil puzzle into an unprecedented opportunity for wealth creation and
common security.
That opportunity rests on a startling fact proven in the next 101 pages:
most of the oil now used in the United States (and the world) is being
wasted, and can be saved more cheaply than buying it. To assess that
solution, let’s start with a common understanding of the problem.
Oil is the lifeblood of
modern industrial economies—but not forever
The United States of America has the mightiest economy and the most
mobile society in the history of the world. Its mobility is 96%15 fueled by
oil costing a quarter-trillion dollars a year16 and consuming seven of every
ten barrels the nation uses. Oil provides 40–43% of all energy used by
the United States, Europe, Asia, Africa, and the world. Oil dependence17
varies—30% in China, 50% in Japan, 59% in Central and South America—
but it’s high everywhere. The whole world is happily hooked on convenient, transportable, versatile, ubiquitous, cheap oil.
Cheap oil,
the world’s seemingly
irreplaceable
addiction,
is no longer the only
or even the cheapest
way to do its vital
and ubiquitous tasks.
14. Wrynn (1993) states at p. 52 that by war’s end, “the automobile industry was responsible for 20 percent of the nation’s war production by dollar volume.”
He shows striking examples (pp. 30, 54, 75, 76) of automakers’ ads emphasizing fuel economy to help the war effort and to help civilians stretch rationed
gasoline. And in 1943, GM (p. 39) advertised not just “Victory Through Progress” but also “Progress Through Victory”: “…from what is learned in the stern
test of war are being gathered many lessons to make more bountiful the blessings of the coming peace.”
15. Measured by energy content, not volume; excluding energy (nearly all natural gas) used to run pipelines (2.7% of transportation energy); excluding here
(contrary to the “Hydrocarbon definitions” convention on p. 40, otherwise used throughout this report) the portion of gasoline-blended oxygenates (1.3% of
total oil use) that don’t actually come from petroleum; and excluding the negligible miscellaneous transportation fuels in note 202, p. 36.
16. The U.S. energy statistics in this section and throughout this report are drawn, unless otherwise noted, from U.S. Energy Information Administration (EIA)
2003c, and forecasts from EIA 2004. Primary sources are often documented in Lovins 2003.
17. The fraction of total primary energy use that is provided by oil. Oil import dependence is how much of the oil used has been imported (usually net of oil
exports, unless specified as “gross” imports).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
1
Oil Dependence
Oil is the lifeblood of modern industrial economies—but not forever
The average U.S.
light vehicle each
day burns 100 times
its weight in ancient
plants in the form of
gasoline.
The oil we’re burning in two centuries took hundreds of millions of years
to form. When the Russian chemist D.I. Mendelyeev figured out what it
was, he exclaimed it was far too precious to burn. We’ve been burning it
ever since—ten thousand gallons a second in America alone. Each gallon
of gasoline took eons to form (very inefficiently) from a quarter-million
pounds of primeval plants. Thus the average U.S. light vehicle18 each day
burns 100 times its weight in ancient plants in the form of gasoline.19 Only
an eighth of that fuel energy even reaches the wheels, a sixteenth accelerates the car, and less than one percent ends up moving the driver.20
Only an eighth
of that fuel energy
even reaches the
wheels,
a sixteenth
accelerates the car,
and less than
one percent ends up
moving the driver.
A 20-mile round trip in an average two-ton21 new light vehicle to buy a
gallon of milk burns a gallon of gasoline at about half the milk’s cost.22
The extraordinary global oil industry has made U.S. gasoline abundant,
cheaper than bottled water, a half to a fourth of Europe’s or Japan’s gasoline price.23 We use it accordingly. A comedian’s acid remark, even if it
touches a sensitive nerve today, could as well have been made under at
least six of the past seven Administrations:24
Two dollars a gallon to go ten miles is too much, but five to the parking valet to
go ten feet is okay. The irony is [that] what we love most about our cars—the feeling of freedom they provide —has made us slaves. Slaves to cheap oil, which has
corrupted our politics, threatened our environment, funded our enemies….Faced
with our addiction to oil, what does our leadership say? Get more of it! Strange
when you consider their answer to drug dependence is to cut off the supply.
The world consumes a cubic mile of oil per year. This is growing by just
over one percent per year and is forecast to accelerate. A third of the
growth supplies number two user China,25 whose car sales soared 56% in
2003.26 By 2025 its cars could need another Saudi Arabia or two. Just one18. “Light vehicles,” sometimes called “light-duty vehicles,” comprise cars, light trucks (sport-utility vehicles [SUVs], pickup trucks, and vans), and
“crossover vehicles” (a new category combining SUV with sedan attributes), with a gross vehicle weight not exceeding 10,000 pounds (4,537 kg).
19. Dukes (2003), adjusted from his assumed 0.67 refinery yield of gasoline to the actual 2000 U.S. average of 0.462 (EIA 2001a, Table 19) and using EIA’s 2.46
kgC/gal for gasoline and carbon contents of 0.855 for crude oil and 0.866 for gasoline (EIA 2002a, p. B-8, Table B-6). The average U.S. light vehicle in 2000
burned 591 U.S. gallons (2,238 L) of gasoline (ORNL 2002, Tables 7.1, 7.2) made from ~65,000 metric tonnes of ancient plants, using the adjusted Dukes coefficient of 111 T/gal. The average new light vehicle sold in 2000 weighed 3,821 lb (EPA 2003). Annual ancient-plant consumption is thus ~37,000× its curb weight,
because, as Dukes explains, only a tiny fraction of the plants ends up in oil that’s then geologically trapped and ultimately recovered.
20. See Box 5 on pp. 39–42.
21. The average Model Year 2003 light vehicle, at 4,021 pounds, “broke the two-ton barrier for the first time since the mid-1970’s”: Hakim 2004a.
22. Michael Lewis, head of Deutsche Bank’s commodity research in London, notes that if Safeway shoppers in Maryland bought a barrel of milk, it’d cost
$138; orange juice, $192; Evian mineral water, $246. By comparison, $40 oil looks cheap. The Wall Street Journal ’s M.R. Sesit (2004) neatly concludes, however: “But oranges grow on trees; oil doesn’t. SUVs don’t run on mineral water. And, Mr. Lewis acknowledges, ‘Nobody’s blowing up cows.’”
23. In 2000, 22 countries sold gasoline at below a nominal benchmark of retail cost, 25 at between cost and U.S. prices, and 111 at higher-than-U.S. prices.
Post-tax prices varied from Turkmenistan’s 8¢/gal to Hong Kong’s $5.53/gal (including $4.31 tax, vs. U.S. tax of 57¢). The U.S. tax rate of 32% of post-tax retail
price was half the OECD average of 67%; in absolute terms, the U.S. tax, pretax price, and post-tax price were respectively 26%, 128%, and 56% of the OECD
average. Diesel fuel was generally taxed substantially less per gallon than gasoline (Bacon 2004). See also Metshies 1999.
24. Maher 2002, pp. 29–30.
25. IEA 2003a, p. 13; average of 2001 through projected 2004 global demand growth; updated by Mallet 2004. China’s ascendancy to number two was seven
years earlier than forecast. In the first four months of 2004, China imported 33% more oil than a year earlier (CNN 2004).
26. People’s Daily (Beijing) 2004; Auffhammer 2004; NAS/NRC/CAE 2003; cf. Wonacott, White, & Shirazou 2004 (~80% growth 2002, 36% 2003).
2
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil is the lifeblood of modern industrial economies—but not forever
eighth of the world’s people own cars; more want one. Africa and China
have only the car ownership America enjoyed around 1915.27
Yet it’s the superaffluent United States whose growing call on a fourth of
the world’s oil is the mainspring of global demand. During 2000–25, oil
use is officially forecast to grow by 44% in the United States and 57% in
the world.28 A fifth of that global increase is to fuel the U.S., which by 2025
would use as much oil as Canada, Western and Eastern Europe, Japan,
Australia, and New Zealand combined. As the richest nation on earth, we
can afford it. But five, soon seven, billion people in poor countries, whose
economies average more than twice as oil-intensive,29 want the same oil to
fuel their own development. The forecast 2000–25 increase in U.S. oil
imports exceeds the 2001 oil use of China and India (plus South Korea).
Competing with those emerging giants can’t be good for their vital development, for fostering their hoped-for friendship and cooperation, or for
the prospects of a peaceful and prosperous world community.
Oil, the world’s biggest business, bestrides the world like a colossus.30
Surely such a stupendous source of energy is indispensable, its cost a
necessity of sustaining modernity and prosperity. Surely its possible substitutes are too small, slow, immature, unattractive, or costly to offer realistic alternatives to rising oil consumption—unless, perhaps, forced down
our throats by draconian taxes or intrusive regulation. And surely it’s premature to speculate about life after oil: 31 let our grandchildren, or someone else’s grandchildren, do that.
This study tests all these comfortable assumptions and finds them
unsound. On the contrary, rigorously applying orthodox market economics and modern technologies, it proves that the services Americans get from
oil could be more cheaply provided by wringing more work from the oil we use
and substituting non-oil sources for the rest. And because the alternatives to
oil generally cost less and work better, they can be implemented quickly
in the marketplace, with all its free choice, dynamism, and innovation.
If guided more by profit and less by regulation and subsidy, this
approach could even help to make government leaner, more flexible,
and more valuable.
Oil Dependence
The services
Americans get from
oil could be more
cheaply provided by
wringing more work
from the oil we use
and substituting
non-oil sources
for the rest.
27. Greene 2004. Demand
growth for personal mobility becomes obvious in the
vast world market only for
big countries with a burgeoning middle class. But
for countries at the bottom
of the development ladder,
oil is for most people an
unimaginable luxury: destitute Chad’s nine million
people, with just 80 miles of
paved roads, use oil at only
the rate of two jumbo jets
making a daily transatlantic
round-trip. (“A 747-400 [the
newest, most efficient
jumbo] that flies 3,500
statute miles [5,630 km] and
carries 126,000 pounds
[56,700 kg] of fuel will consume an average of five
gallons [19 L] per mile.”
[Boeing, undated] The nominal flight distance New
York to London is 3,461
statute miles, so a round
trip uses ~824 barrels,
unadjusted for refining.
Chad used ~1,500 bbl/d in
2002 [www.theodora.com
2003]). If rich countries use
too much oil, Chad can
scarcely afford any. But all
this could change now that
Chad has discovered oil—
and signed a pathfinding
transparency agreement
guiding its development.
28. EIA 2004b.
29. IEA 2003, p. 74. The IEA’s May 2004 update (IEA 2004b) finds that oil price rises are twice as damaging to the Asian economy as to OECD’s, four times in
“very poor highly indebted countries,” and at least eight times in sub-Saharan African countries.
30. Oil economist and World Bank consultant Mamdouh G. Salameh (2004) summarizes: “One could not imagine modern societies existing without oil….Oil
makes the difference between war and peace. The importance of oil cannot be compared with any other commodity because of its versatility and dimensions, namely, economic, military, social, and political. The free enterprise system…and modern business owe their rise and development to the discovery of
oil. Oil is the world’s largest and most pervasive business….Of the top 20 companies in the world, 7 are oil companies.”
31. Roberts 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
3
Oil Dependence
Just as whale oil was
outcompeted before
it was depleted,
powerful technologies
and implementation
methods could make
oil uncompetitive even
at low prices before
it becomes unavailable
even at high prices.
Even an important industry
can be displaced by competitors
Discovering such a startling, unseen, seemingly radical possibility right
under our noses is not a pipe dream; it’s a normal event in the history of
technology, as Figure 1 illustrates.
In whaling’s 1835–45 heyday, millions of homes used the clean, warm,
even light of sperm-oil lamps and candles. As sperm whales got scarcer,
the fleet hunted more plentiful but less oil-rich species whose inferior oil
fetched half the price. But meanwhile, whale-based illumination’s price
had stayed high enough for long enough to elicit two fatal coal-based
competitors (Technical Annex, Ch. 1): “town gas” and “coal-oil” kerosene.33
As both steadily spread, whaling peaked in 1847. Ten years later, Michael
Dietz’s clean, safe, smokeless, odorless kerosene lantern was imported
and promptly entered cheap U.S. production. In just three more years,
whale oil’s disdained competitors became dominant: the whalers lost
their customers before they ran out of whales, as the “public abandoned
Figure 1: Rise and fall of the U.S. whaling fleet, 1821–84
Whale oils already were uncompetitive and losing share before populations crashed and Drake struck oil.32
1881
1876
1871
1866
1861
1856
0
1851
crude-oil production (100,000 gal/y)
20
1846
crude-oil wellhead price (2000 $/gal)
1841
whale oil price (2000 $/gal)
40
1836
whaling fleet displacement (10,000 tons)
60
1831
sperm oil price (2000 $/gal)
80
1826
sperm oil production (100,000 gal/y)
100
1821
whale oil production (100,000 gal/y)
various units (see legend)
120
year
Source: RMI analysis based on Goode 1887 (note 32), U.S. Bureau of the Census 1975.
32. Goode 1887, pp. 168–173. CPI (Consumer Price Index) deflator from McCusker 1991, completed for 1999–2000 using CPI–U (CPI for All Urban Consumers)
of the U.S. Department of Commerce, Bureau of Economic Statistics. A GDP implicit price deflator was also kindly provided by Dr. Philip Crowson, calculated
from U.S. Bureau of Mines 1993 at pp. 51–52 (copper) and 201, but was not used here because that series goes back only to 1870, and the near-doubling of
consumer prices during 1860–65 makes earlier deflators relatively unreliable. All the real prices graphed here should therefore be taken as indicative but not
necessarily accurate. Petroleum data (deflated with same McCusker CPI): USCB 1975.
33. Camphene (from ethanol and turpentine), lard oil, various mixtures, and tallow candles were also lesser competitors. Camphene was the most important;
it gave good light but sometimes caused accidents. It had about a tenth of the market for the main illuminants and lubricants until 1862–64, when a $2.08/gal
alcohol tax to help fund the Civil War, meant to apply to beverage alcohol but written to tax all alcohol, threw the lighting market to kerosene. Davis, Gallman,
& Gleiter 1997, pp. 365–366; Kovarik 1998.
4
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Even an important industry can be displaced by competitors
whale oil almost overnight.” 34 In the 1850s, whale oils lost two-thirds of
their total market share (by value), even more if lubricants are factored
out.35 By the end of the decade, whale-derived illuminants sold under
$3 million,36 while coal-derived kerosene sold $5 million in 1859,37 and
221 town-gas networks, up from 31 in 1850,38 sold $12 million in 1860.39
But also in 1859, Drake struck oil in Pennsylvania, making kerosene ubiquitous within a year, and by 1865, a third to a fourth the price of sperm
or whale oil per unit of energy.40 Ultimately, town gas got replaced in turn
by natural gas; gas and kerosene lights, by Edison’s 1879 electric lamp.
But by the 1870s, the whaling industry was already nearly gone, pleading
for federal subsidies on a national-security rationale.41 The remnant whale
populations had been saved by technological innovators and profit-maximizing capitalists.42
Around the mid-1850s, an astute investor with no gift of prophecy could
have foreseen that the better and cheaper substitutes on the market
and in inventors’ laboratories could quickly capture whale oil’s core market. Oddly, nobody at the time seems to have done a clear-eyed assessment adding up whale oil’s competitors, so the industry didn’t see them
coming until too late.
A century and a half later, history looks set to repeat itself in the petroleum
(“rock-oil”) industry. Its mature provinces are in decline, more-volatile
prices show new upward bias, and exploration is being driven to remote,
costly, and hostile frontier provinces, just as in this 1887 comment: 43
The general decline of the whale-fishery, resulting partly from the scarcity of
whales, has led to the abandonment of many of the once famous grounds,
and cargoes of sperm oil are obtained only after the most energetic efforts in
scouring the oceans.
Oil Dependence
34. San Joaquin Geological
Society 2002. As Doug
Koplow points out, the transition was quick because it
took only individual action,
not collective action like
creating a town-gas network. Lamp expert Heinz
Baumann (personal communication, 1 June 2004)
notes that lamps designed
for any fuel—whale oil,
camphene, etc.—were
“commonly converted” to
the superior kerosene product, facilitated by de facto
standardization to three
thread sizes. Kerosene
providers often offered the
needed conversion kit.
It cost ~$0.5–1 in nominal
dollars, comparable to a
gallon of kerosene ($0.59 in
1859), whale oil ($0.49 but
inferior), or $1.36 (sperm
oil), and typically paid back
in months.
35. Davis, Gallman, & Gleiter 1997, p. 362.
36. Goode (1887) shows $5.4 million in total whale-and-sperm-oil sales in 1860, of which, however, $3.2 million was sperm oil almost wholly committed to
lubricant and export markets (Davis, Gallman, & Gleiter 1997, p. 345).
37. Yergin 1991, p. 23. This 1859 comparison is in nominal 1859 dollars.
38. Tarr 1999, pp. 19–37; Tarr 2004. In 1850, manufactured gas had one-fourth the value of whale, sperm, and fish oils; in 1860, twice their value (Davis,
Gallman, & Gleiter 1997, pp. 353–4).
39. Davis, Gallman, & Gleiter 1997, p. 353, citing the 1865 Eighth Census of Manufactures for 1860.
40. Kerosene sold for $0.59/gal nominal in 1859 (Robbins 1992), or $6.20/gal in 2000 $. A gallon of sperm oil has the same energy content (~41.9 MJ/kg) as a
gallon of light petroleum, or 91% that of kerosene (Hodgman, Weast, & Selby 1961, pp. 1945, 1936); we assume whale oil does too.
41. Starbuck 1878/1989, p. 113.
42. The authors have been unable to find time-series data for the production or consumption of U.S. town gas or coal-oil before 1859, but these fuels clearly
won before Drake struck oil: Robbins 1992; Davis, Gallman, & Hutchins 1988; Davis, Gallman, & Gleiter 1997, pp. 342–362, 515.
43. Starbuck 1878/1989. “The increase in [human] population would have caused an increase in [whale-oil] consumption beyond the power of the fishery to
supply, for even at the necessary high prices people would have had light. But…[t]he expense of procuring [whale-]oil was yearly increasing when the oilwells of Pennsylvania were opened, and a source of illumination opened at once plentiful, cheap, and good. Its dangerous qualities at first greatly checked
its general use, but, these removed, it entered into active, relentless competition with whale-oil, and it proved the more powerful of the antagonistic forces.”
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
5
Oil Dependence
Even an important industry can be displaced by competitors
Techniques for finding and lifting oil
have made stunning
progress too.
But innovations to
save and replace oil
are progressing even
faster. Oil thus has
dwindling resources
and rising costs;
its competitors have
expanding resources
and falling costs.
Fortunately, the power of today’s technologies is immeasurably greater
than that of 1887. The modern competitors that could capture petroleum’s
key markets range from doubled-efficiency hybrid-electric cars to ultrathin heat insulation and superwindows, from windfarms and asphalt substitutes to advanced aircraft. Information systems hook up the world,
streamline business operations (design, process control, freight logistics,
building management), and optimize everything with cheap microchips.
Laboratories are hatching such marvels as doubled-efficiency heavy
trucks, competitive solar cells, liquid-hydrogen cryoplanes, and quintupled-efficiency carbon-fiber cars powered by clean hydrogen fuel cells.
To be sure, techniques for finding and lifting oil have made stunning
progress too. But innovations to save and replace oil are progressing even
faster. Oil thus has dwindling resources and rising costs; its competitors
have expanding resources and falling costs. Shrewd investors are starting
to hedge bets and re-allocate assets.
44. Grübler, Nakićenović,
& Victor 1999.
45. Abernathy 1978, pp.
18–19, 65, 183–5. The
closed-body price premium
fell from ~50% in 1922 to 5%
in 1926 (p. 19).
46. By April 2004, more people owned cellphones in
China (296 million) than
there are people in the U.S.
(293 million), and China was
adding two subscribers per
second—even though only
a fifth of Chinese people yet
had cellphones, vs. half in
the U.S., two-thirds in
Japan (Bloomberg 2004),
and >100% saturation in
Hong Kong and Taiwan.
The world’s 1.2 billion cellphones in 2001, one for
every 5.2 people, are
expected to become 2 billion by 2006, one for every
3.3 people. By 2002, wireless telephony was a global
half-trillion-dollar-a-year
business. In Germany, not
traditionally a hotbed of
innovative telecom
providers, the fraction of
the population with cellphone subscriptions doubled to nearly 60% in the
single year 2000.
6
Nobody knows how quickly these new oil-displacing techniques will
spread, but to a great degree that speed is not fate but choice. Transitions
can be swift when market logic is strong, policies are consistent, and institutions are flexible. It took the United States only 12 years to go from 10%
to 90% adoption (in the capital stock, not new sales) in switching from
horses to cars, from uncontrolled automotive emissions to catalytic converters, and from steam to diesel/electric locomotives; 15 years from vacuum-tube to transistor radios and from black-and-white to color televisions.
Even such a big infrastructure shift as intercity rail to air travel took just
26 years.44 U.S. autobody manufacturing went from 85% open and made
of wood in 1920 to more than 70% closed and made of steel just six years
later.45 Techniques available to any individual, like cellphones, tend to
spread fastest.46
The powerful portfolio of ways to accelerate oil efficiency and non-oil
supplies isn’t only technological: innovations in implementation are
equally rapid and important, and they all reinforce each other. Nations
can speed up spontaneous market transitions by turning purpose into
policy. Creative, common-sense frameworks can clear away decades of
underbrush and obstacles, solve problems at their root cause, and command wide support across ideological boundaries. Global challenges are
starting to reinspire the public engagement that empowers citizens to
manifest their patriotism in their everyday choices. Major sectors of the
economy need the revitalization and job growth that this agenda can
deliver. If Americans, with business in the vanguard, chose to mobilize
their resources—and to help make markets more mindful of new profit
potential—the world’s premier market economy could pivot and charge
through obstacles as unstoppably as football legend Walter Payton.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Even an important industry can be displaced by competitors
Oil Dependence
By combining new technologies, business strategies, marketing methods,
policy instruments, and community and personal choices, we can displace
more oil, faster, cheaper, than ever imagined. This report shows how.
And there’s a precedent.
America can replace oil quickly—
and already has
The United States last paid attention to oil during 1977–85, spurred by
the 1979 “second oil shock,” which raised prices even more than the 1973
Arab embargo had done. In those eight years, the United States proved it
could boost its oil efficiency faster than OPEC could cut its oil sales: the U.S. had
more flexibility on the demand side than OPEC had on the supply side. While
U.S. GDP rose by 27%, oil consumption fell by 17%, net oil imports fell
by 50%, and net oil imports from the Persian Gulf fell by 87%.47 The drop
in oil use was three times 1977 imports from Saudi Arabia, nearly twice
today’s imports from the Gulf. Faced with suppliers’ “oil sword,” we
drew our own and won the day as plummeting American imports took
away one-seventh of OPEC’s market. The entire world oil market shrank
by one-tenth, OPEC’s global market share was slashed from 52% to 30%,
and its output fell by 48%, breaking its pricing power for a decade.48
During 1977–85, U.S. oil intensity (barrels per dollar of real GDP)
dropped by 35%. If we’d resumed that pace of oil savings—5.2% a year—
starting in January 2001, the United States could have eliminated the
equivalent of its 2000 Persian Gulf imports by May 2003 at constant GDP,
or by 2007 with 3%/y GDP growth. And we could rerun that old play,
only better—starting, as we did then, with our personal motor vehicles.
The cornerstone of the 1977–85 revolution in U.S. oil savings—when
Washington led with coherent policy and Detroit rose to the occasion—
was 7.6-mpg-better domestic cars. On average, each new car drove 1%
fewer miles on 20% fewer gallons, achieving 96% of that efficiency gain
from smarter design, only 4% from smaller size.49 During 1975–84, the
fuel economy of the entire light-vehicle fleet rose by 62% while vehicles
became safer, far cleaner, and no less peppy.50 Detroit then kept on innovating, but once its success had crashed the world oil price in 1985–86,
ever-better powertrains were used to make cars more muscular, not more
In the eight years
when we last paid
attention to oil
(1977–85), GDP grew
3% a year, yet oil use
fell 2% a year.
That pace of reducing
oil intensity, 5.2%/y,
is equivalent, at a
given level of GDP,
to displacing a Gulf’s
worth of oil every
two and a half years.
47. EIA 2003c shows gross
imports from the Gulf as
2.448 Mbbl/d in 1977 and
0.311 (0.309 net) in 1985. The
gross-minus-net-imports difference—exports from the
U.S. to the Gulf—was only
0.002 Mbbl/d in 1985, averaged 0.0034 Mbbl/d (0.54%)
during 1981–85, and is
unpublished and apparently
unavailable from EIA for all
years before 1981. Plausibly
applying the 1981–85 average export rate to 1977–80
yields inferred 1977 net
imports of 2.448 – 0.003 =
2.445 Mbbl/d, falling 87.4% to
1985’s 0.309 Mbbl/d. By mentioning Persian Gulf imports
in this report, we don’t mean
to imply this is a critical
index; more important is
how much of its oil the U.S.
imports and how much of
world supply (and of world
traded supply) comes from
the Persian Gulf.
48. The U.S. contribution was vital but not unique, and was part of longer-term trends throughout the industrialized world. During 1973–2002, oil as a fraction
of total primary energy fell from 80% to 48% in Japan, 64% to 33% in Europe, and 48% to 37% in the world, but only from 47% to 39% in the U.S., which in
these terms did worse than the global average. Even during 1979–85, the golden age of U.S. oil savings, oil’s share of primary energy fell only from 45% to
40% in the U.S., vs. 72% to 55% in Japan, 55% to 45% in Europe, and 45% to 38% in the world: Franssen 2004.
49. Patterson 1987: “If the 1976 size class shares for autos were applied to the 1987 car class fuel economies, the resulting new car MPG would be 27.7 in 1987
(just 0.4 MPG less than the actual values). Thus, if in 1987 the nation had reverted back to the 1987 new car size mix, the eleven year gain of 10.9 MPG would
have been reduced by only 4 percent.” This is not valid for light trucks, whose size shift was dominated by sale of smaller imports to new pickup buyers.
50. NAS/NRC 1992, Fig. 1-1; ORNL 2003; EPA 2003.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
7
Oil Dependence
America can replace oil quickly—and already has
Oil
(or more precisely,
the service provided
by oil)
makes us strong,
but oil dependence
makes us weak.
frugal. The average new U.S. light vehicle in 2003 had 24% more weight,
93% more horsepower, and 29% faster 0–60-mph time than in 1981, but
only 1% more miles per gallon. If 1981 performance had instead stayed
constant, light vehicles would have become 33% more efficient,51 displacing more than` 2000 Persian Gulf imports. America’s light-vehicle fleet
today is nearly the world’s most fuel-efficient per ton-mile, but with more
tons, it uses the most fuel per mile of any advanced country. It needn’t.
We’ll see how the same engineering prowess, harnessing a whole new set
of advanced technologies, could make that fleet the world’s least fuelintensive52 while continuing to offer all the sizes and features that customers want.
Which would be
worse—if Saudi
Arabia couldn’t meet
huge forecast
increases in output,
or if it could?
One attack on a key
Saudi oil facility could
crash the world economy at any moment.
American business
is particularly
vulnerable to
oil disruptions.
Yet the rising dangers
of dependence on
the world’s most
vulnerable sources,
both abroad and at
home, are a selfinflicted problem.
We can profitably
eliminate it by
choosing the best buys
first.
Similar opportunities beckon throughout the U.S. economy. To understand their true value, let’s review how, as security scholar Michael Klare
puts it, oil (or more precisely, the service provided by oil) makes us
strong, but oil dependence makes us weak.
Oil supplies are becoming more concentrated
and less secure
The imported fraction of U.S. oil is officially projected to rise from 53% in
2000 to 70% in 2025 (Fig. 2), vs. 66% in the E.U. and 100% in Japan. By 2025,
U.S. imports are projected to rise by half, while imports from OPEC’s 11
member nations and from the Persian Gulf are nearly to double.
In 2000, 2.5 million barrels per day (Mbbl/d53) and nearly half of world oil
exports came from the Persian Gulf, an unstable region racked by violent
ethnic, religious, and political conflicts.54 The Gulf provided 24% of U.S.
net imports or 13% of total oil consumption. The 1980 Carter Doctrine
declared that “any attempt by an outside force to gain control of the
Persian Gulf will be regarded as an assault on the vital interests of the
United States of America, and…will be repelled by any means necessary,
including military force.” U.S. forces have since engaged in at least four
51. EPA 2003. Roberts (2004, p. 155) states that “of the nearly twenty million barrels of oil that America uses every day, more than a sixth [actually 13% in 2000]
represents a direct consequence of the decision by automakers to invest the efficiency dividend in power, not fuel economy.” That remains true regardless of
whether automakers were passively responding to the demands of sovereign consumers or designing their marketing to influence those demands.
52. Schipper 2004: “…[A]utomobiles in the United States actually use less energy per kilometer and per unit of weight than do those in nearly every other
country. By that measure, U.S. cars are efficient. But they also use more energy per kilometer because they weigh more than cars elsewhere. Thus, one can
say that the U.S. car fleet is more fuel intensive than fleets in any other advanced countries. The reason is that American behavior ‘favors’ large cars, largely
due to low car prices and low fuel prices”—and to the habit, no longer necessary (see pp. 53–67), of making them out of heavy steel.
53. Throughout this report, as noted in Box 5 (“Conventions”), “bbl” means barrel (42 U.S. gallons) and “M” is used in its scientific sense of “million” rather
than in its engineering sense of “thousand.” We therefore avoid the ambiguous usages (common in the energy literature) of Mbpd, MMBTU, etc.
54. More than 90 conflicts in the whole Middle East killed 2.35 million people between World War II and sometime in the early 1990s: Delucchi & Murphy
1996, p. 2, note 3, citing Cordesman 1993, pp. 5–8.
55. Protecting U.S.-reflagged Kuwaiti tankers in the Iran-Iraq war of 1980–88; the 1991 Gulf war; the subsequent containment of Saddam Hussein; and the
2003 Iraq war. This doesn’t include the 2001 Afghanistan war or many other skirmishes. Klare (2004) notes that most of Central Command’s casualties have
been incurred in the Gulf and in its well-documented primary mission—ensuring access to Middle Eastern, chiefly Persian Gulf, oil.
8
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Oil supplies are becoming more concentrated and less secure
Figure 2: U.S. net oil imports and sources in 2000 and (approximately) in EIA’s 2025 projection
2000 net oil imports by country
(10.42 Mbbl/d)
15
Sa %
ud
iA
rab
ia
15%
Vene
zuala
5.28
Gu
lf
9.57
9.33
8.62
2000
(19.75 Mbbl/d)
2025
(28.3 Mbbl/d)
Am
er
ic
a
a
eri
Nig
Algeria 2%
Caribbe
an 12%
9%
Ca
nad
a1
6%
o 10%
Mexic
4.85
2.49
2.65
24
%
er
9% oth Gulf
n
Persia
y 7%
% n
2 5 r s ia
Pe EC
OP
24
o th %
er
OP
EC
No
rth
6%
Far East 2%
orw a
n
5.26
Europe 2%
in/N
be
a
2025 gross oil imports by region
(20.7 Mbbl/d)
%
11
er
oth
Brita
r ib
%
other 5
3%
o la
Ang a 3%
bi
lom
Co
Ca
Persian Gulf
non-Gulf OPEC
non-OPEC
domestic
Since EIA projects only gross (not net) 2025 imports by region, we approximate the distribution
between red, yellow, and green for 2025 by using EIA 2025 gross regional imports, minus EIA 2025
exports prorated to these regions according to 2000 data (when 98 percent of U.S. oil exports went
to non-OPEC, 1.75 percent to non-Gulf OPEC, and 0.25 percent to Gulf OPEC). The bar charts show
refined products and LPG supplied, including oxygenates, blending agents, and refinery gains.
Above graph shows gross imports, from which 1.02 Mbbl/d of
exports with unknown regional distribution would be subtracted
to arrive at the 19.68 Mbbl/d of net imports shown in the 2025 bar
chart. The Caribbean share of gross imports for 2025 correctly
represents EIA data (but not the graph on p. 68 of AEO04) and
may represent products refined in the Caribbean from crude oil
of various origins.
Source: (left) EIA 2001a, EIA 2004, EIA NEMS Database ; (right) EIA 2004, p. 68, fig. 44.
conflicts there55 without stabilizing the region. Moreover, despite intensive
worldwide efforts to find alternative oil sources, the Gulf’s ~65% share of
declared global reserves (and even more of the cheapest oil) is up from
54% two decades earlier.56 Thus the U.S. depends ever more on the
region57 that’s most volatile, militarized, geopolitically challenged, and
hostile to American values.
Today the U.S. imports twice as much oil as it did in 1973, when a hiccup
in one-eighth of the supply doubled unemployment, slashed 1975 GDP by
3–5.5%, and quadrupled oil prices in weeks. Prolonged shortages could rip
the fabric of American society, throttling everything from daily commuting
and air travel to food trucking.58 And that could actually happen.
56. BP 2003, pp. 4–5.
Saudi Arabia, the world’s sole “swing producer” (with spare capacity to
ramp up if prices soar), holds one-fourth of global oil reserves and ~80–
85% of spring 2004 spare output capacity,59 and provided 63% of U.S. net
imports from the Gulf in 2000. Saudi Arabia’s fractious, fragile monarchy
faces a “slow-motion insurrection” 60 and rising terrorism “focused on the
58. The average molecule
of American food is said to
travel ~1,400–1,500 miles,
mainly by truck, before it’s
eaten. If trucking stopped
(or a few bridges across
the Mississippi went down),
the East Coast could run
short within days.
57. Its complexities and heterogenities are surveyed by
Cordesman 2004.
59. Bahree & McKay (2004) report that although Saudi Arabia could (barely) offset “a sharp drop in output from Iraq, which is producing an estimated 2.5 million barrels a day,…the world’s oil exporters are mostly pumping flat out and, taken together, couldn’t make up a loss of even a 15% reduction in output, analysts say.” Russia was hoped to become a second swing producer, but this is now in doubt, and the two countries may even be increasing their cooperation
more than becoming independent counterweights to each other. The ~80–85% estimate of the Saudi share of world crude-oil swing capacity is from EIA
2004c. A year earlier, most estimates were around two-thirds. Unfortunately, most Saudi crudes are too sour (high in sulfur) for many developed-country and
Chinese refineries, so the 2004 shortage of sweet crudes is more acute than the aggregate supply/demand balance implies.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
9
Oil Dependence
Oil supplies are becoming more concentrated and less secure
60. Dyer (2004) says a 2001
secret poll by the Saudi
Interior Ministry found that
95% of Saudi males aged
19–34 approved of bin
Laden’s attacks on the U.S.,
whose alliance with and
presence in the Kingdom
has exacerbated severe
underlying economic and
demographic pressures.
two pillars of the Saudi relationship61 with the West: energy and defense.”62
About half of Saudi oil capacity comes from one oilfield. Two-thirds of
Saudi oil goes through one processing plant and two terminals, the larger
of which was the target of a foiled Saudi Islamist attack in mid-2002.63
Simple attacks on a few key facilities, such as pipeline nodes, could choke
supply for two years.64 The defenders can’t always win. Anticipating
trouble, the global oil industry has achieved impressive supply diversification, routing flexibility, stockpiling, hedging, and other precautions,
including International Energy Agency (IEA) member nations’ stockpiles65
(nearly half in the U.S. Strategic Petroleum Reserve66). Yet the IEA’s former Chief Economist concludes that “reliance on the sole Saudi pillar” of
world oil-market stability “will continue”—and that a few months’ halt in
Saudi exports “would spell disaster” and “throw the global economy into
chaos,”67 even if nothing else were harmed.68
61. Ever since King Abd alAziz [Ibn Saud] struck a
bargain with President
Roosevelt in 1945, Saudi
Arabia, or more precisely
the Saudi royal family, has
been an American protectorate in all but name—
shielded by American military might from external and
internal enemies in
exchange for selling oil
(except in the 1973–74
embargo). The intricate history, including a failed 1943
U.S. effort to buy the major
Saudi oilfields outright
through the Petroleum
Reserves Corporation, is
summarized by Klare 2004
and more fully reviewed by
Pollack 2002.
That’s a big if. The world’s key oil terminals, shipping lanes, ports,
pipelines, refineries, and other facilities could be devastated by plausible
small-group attacks.69 Oil facilities are routinely attacked in Iraq, Colombia,
Ecuador, Nigeria, and Russia, and are buffeted by political turmoil in
Venezuela, Iran, and Nigeria. They fuel rivalries and secessionism in
Indonesia, Sudan, the Caspian, and Central Asia. The next act in this
drama may be attacks in three key marine straits (Hormuz, Malacca, and
Bab el-Mandab), perhaps in league with local pirates armed with rocketpropelled grenades and even ship-to-ship missiles.70 FBI Director Mueller
62. Kerr (2004), quoting Kevin Rosser of Control Risks Group, referring to the 1 May 2004 attack on ABB Lummus’s Yanbu office. These attacks were clearly
meant to spark an exodus of the 6 million expatriates, including ~35,000 Americans, on whom the Saudi oil industry heavily relies. Many have since left.
The murder of 22 people at Khobar on 29–30 May 2004 and subsequent attacks have speeded that exodus (MacFarquhar 2004).
63. Luft & Korin 2003; Bradley 2004; Arabialink.com 2002; Lumpkin 2002.
64. Hersh 2001. See also Baer 2003 and Baer 2003a.
65. Financial managers, tempted by mature spot and futures markets and wrongly believing that financial hedges and physical inventories are equivalent,
have tended to wring out private buffer stocks as a carrying cost while hedging in futures and options markets. This further burdens the inadequate public
stockpiles. Nonetheless, the total of all public and private stockpiles worldwide is probably around four billion barrels (Warren 2004). Some of this is required
to keep oil systems operating. Stockpiles of refined products must be changed periodically to prevent spoilage.
66. This 0.7-billion-barrel-capacity underground reservoir complex in the Gulf of Mexico was 94% full in May 2004, aiming at 100% by summer 2005. Its 0.65
billion barrels, available on two weeks’ notice, are equivalent to 58 days of 2003 net imports, but its maximum drawdown rate is only three-eighths as big as
total net imports, and can deliver 4.3 million bbl/d, or 173% of 2003 Gulf gross imports, for an initial 90 days; the other two-fifths of its capacity has only a 1
Mbbl/d delivery rate. Its facilities and pipeline links to refineries are highly centralized and could be destroyed by a small group of saboteurs.
67. Franssen 2004.
68. Besides ~2.6 billion barrels of private stocks, not all usable due to operational constraints, IEA public emergency stocks of 1.4 billion bbl can produce 12.8
Mbbl/d for a month, or 8 Mbbl/d for three months, or 3 Mbbl/d for five months. The biggest historical supply disruption, in 1978–79, was 5.6 Mbbl/d for six
months. In 2002, exports were 7 Mbbl/d from Saudi Arabia and totaled 15.5 Mbbl/d from the Persian Gulf (EIA 2003a). By May 2004, Saudi output had risen to
an estimated 9.1 Mbbl/d and Saudi spare capacity had shrunk to ~1.5 Mbbl/d (both quite uncertain). China kept little stockpile but is starting to create a more
robust one, with a goal of 22 million tonnes (~161 Mbbl) by 2010 (CNN 2004). Until China and other Asian nations have large stocks, IEA stocks and shortagesharing arrangements could be badly stressed.
69. Lovins & Lovins 1982: a 1981 report to DoD, 436 pages, 1,200 references.
70. Pirates, often from international criminal syndicates, have attacked and plundered 96 ships along Africa’s coast; 22% of pirate attacks on worldwide
shipping in 2003 were on tankers (Hosken 2004). The coordinated 24 April 2004 attacks on Iraq’s two offshore oil terminals by dhow and speedboat suicide
bombers are further worrisome illustrations of the risk.
10
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Oil supplies are becoming more concentrated and less secure
About half of Saudi oil capacity comes from one oilfield.
reports that “any number of attacks
71
Two-thirds of Saudi oil goes through one processing plant and two
on ships have been thwarted,”
terminals, the larger of which was the target of a foiled Saudi Islamist
but in 2003 alone, about 100 tankers
attack in mid-2002. Al Qa‘eda calls oil the “umbilical cord and lifeline
were attacked. Some hijacked vesof the crusader community.”
sels became unregistered “ghost
ships.” 72 The Strait of Hormuz is now targeted by anti-ship missiles
emplaced by Iran’s mullahs, perhaps to discourage awkward inquiries
into their apparent nuclear-weapons program.
Oil facilities attract terrorists,
A handful of people could halt three-fourths of the oil and gas supplies
“undermining the stability of the
to the eastern states in one evening without leaving Louisiana.
regimes they are fighting and economically weakening foreign powers with vested interests in their
region.” 73 Al Qa‘eda calls oil the “umbilical cord and lifeline of the crusader community,” 74 and in April 2004 specifically incited attacks on key
Persian Gulf installations.75 As those attacks began, the market, seeing
Gulf oil “in the crosshairs,” 76 added a ~$5–12/bbl risk premium.77 This
71. Luft & Korin 2003.
may mark the start of a new level of interacting social, economic, and
72. The International
Chamber of Commerce’s
political instabilities that lead global oil into an unpleasant era of anxiety
International Maritime
and chronic disruption. Saudi Arabia, spending 36% of its budget on
Bureau’s annual piracy
report for 2003 reports 469
defense,78 might yet avoid both social upheaval79 and external attacks on
incidents (including 100
its concentrated oil facilities—for example, via aircraft that could be
armed attacks), 21 seafarers killed and 71 missing,
hijacked anywhere in the often security-lax region. Nevertheless a slow,
359 taken hostage, 311
“dripping” kind of sabotage could block the Kingdom’s swing producships boarded, and 19
hijacked (ICC Commercial
tion—the mainstay of liquidity from “the world’s central banker for
Crime Services 2004). The
80
petroleum.” Insider collusion in the 2003 Riyadh bombings and 2004
Bureau recommends adding electrified anti-boarding
Yanbu shooting, Al Qa‘eda efforts to assassinate Saudi security officials,81
barriers (ICC Commercial
82
and weekly sabotage of Iraqi oil facilities all suggest that such scenarios
Crime Services 2004a).
73. Luft & Korin 2003.
74. Luft & Korin 2003.
75. Both the 1 May 2004 shooting spree in Yanbu and the 24 April 2004 dhow-and-speedboat-bombs attack on the Basra oil terminal (which exports ~85% of
Iraq’s oil) were called for in early April by Al Qa‘eda (Sachs 2004). Days before the Basra attack, U.S. intelligence had warned all Gulf states of such attacks
using boats, jet-skis, or explosives hidden in marine shipping containers (Jordan Times 2004; Agencies 2004).
76. Pope 2004. The quotation is from John Kilduff, senior VP for energy risk management at Fimat USA (Société Générale): Banerjee 2004.
77. Cummins 2004; Banerjee 2004a.
78. Karl 2004.
79. Franssen 2004; Baer 2003; Baer 2003a; UNDP 2002/3; McMillan 2001; Sciolino 2001; and Hersh 2001. Cordesman 2002–04 provides extensive background.
80. This phrase is due to Bahree & McKay (2004). Saudi prospects are also questioned because of high (>30%) and increasing water cuts in production from
the immense but mature Al Ghawar field—about half of Saudi production capacity—and analogous issues in other fields: Darley (2004, pp. 16–17) summarizing the stark contrast between the 24 February 2004 Center for Strategic and International Studies presentations by Simmons (2004) and Abdul Baqi & Saleri
(2004); Reed 2004; Bahree 2004. Similar issues are arising elsewhere, e.g., Gerth & Labaton 2004. The popular view that “the Middle East was poorly explored
and has a huge potential” for future giant and supergiant oilfield discoveries may be geologically unsound (Laherrere 2004).
81. Jehl 2003. See also Pope 2004a.
82. Luft (2004a) notes that “an average of one to two sabotage attacks a week against Iraq’s oil pipelines” cut prewar exports of 2.5 Mbbl/d to 1.5 Mbbl/d
despite nearly 14,000 security guards and electronic surveillance equipment. “After more than 100 pipeline attacks in northern Iraq, terrorists last month
began hitting the pipelines in the south near Basra.” (Mid-June pipeline bombings then cut off all Basra exports, costing several hundred million dollars.)
Saudi Arabia, with over 10,000 miles of pipelines, mostly aboveground, has over twice the network size of Iraq. In all, pipelines, “moving 40% of the world’s
oil across some of the world’s most volatile regions,…are a real prize for terrorists.” See also IAGS 2004; Lovins & Lovins 1982; Baer 2003b.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
11
Oil Dependence
Oil supplies are becoming more concentrated and less secure
83. Bahree & McKay 2004.
84–88. See next page.
The United States
produces 21% of
Gross World Product
and uses 26% of the
world’s oil, but
produces only 9%
and owns
a mere 2–3%,
so we can’t
drill our way out of
depletion.
The policy issue
is the level of oil ,
not import , dependence, so more “oil
independence”
is not an optimal goal.
aren’t fanciful.83 Some in Washington continue to muse about seizing the
Kingdom’s oil-rich east if the House of Saud implodes.84 Understanding the
scope for American mobilization to stop needing that oil would seem more
prudent. Executing such a contingency plan could preemptively blunt the
oil weapon that America’s enemies increasingly seek to wield.
Even if the United States imported no oil and its economy weren’t intertwined with others that do, its own oil still wouldn’t be secure. Its infrastructure is so vulnerable that a handful of people could halt three-fourths
of the oil and gas supplies to the eastern states in one evening without
leaving Louisiana.85 Some is as vulnerable as the most worrisome Persian
Gulf sites (Box 1).86
Domestic oil is limited
America’s domestic output has declined for 34 years, returning in 2003 to
its ~1955 level. Prolonged reversal is not a realistic or profitable option.
After 145 years of exploitation, U.S. reserves are mostly played out, so
new oil typically costs more at home than abroad.87 The United States,
with 4.6% of the world’s people, produces 21% of Gross World Product
and uses 26% of the world’s oil, but produces only 9% and owns a mere
2–3% (including all off-limits areas),88 so we can’t drill our way out of
depletion. Over time, other non-Mideast oil areas will follow this pattern
too. But a market economy with relatively limited and costly oil of its own
1: An example of domestic energy vulnerability
The 800-mile Trans-Alaska Pipeline System (TAPS) delivers one-sixth of U.S. refinery input and of
domestic oil output. TAPS is rapidly aging, suffers from thawing permafrost, has persistent and
increasingly serious maintenance and management problems (shown from the accidental destruction
of a key pumping station in 1977 to repeatedly botched recent restarts), and has been shot at more
than 50 times and incompetently bombed twice. It luckily escaped destruction by a competent bomber
caught by chance in 1999, and by a fire averted at the Valdez terminal in 2000. It was shut down for 60
hours in 2001 by a drunk’s rifle shot and for two days at New Year’s 2004 by a terrorism alert. If interrupted for a midwinter week, at least the aboveground half of TAPS’s 9 million barrels of hot oil would
probably congeal into the world’s largest Chap Stick™. TAPS is the only way to deliver potential
reserves beneath the Arctic National Wildlife Refuge (ANWR), the biggest onshore U.S. oil prospect,
estimated by the U.S. Geological Survey to average 3.2 billion barrels—enough to meet today’s U.S. oil
demand for six months starting in a decade. But if Refuge oil were exploited despite what USGS found
to be dismal economics, it would double TAPS’s throughput, making TAPS for several more decades an
even more critical chokepoint than the famously vulnerable Strait of Hormuz.
12
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil supplies are becoming more concentrated and less secure: Domestic oil is limited
can cut dependence on cheaper oil imports in only three basic ways,
of which the first two have been officially favored so far:
• Protectionism taxes foreign oil or subsidizes domestic oil89 to distort
their relative prices. (In the case of subsidies, it also retards efficient
use and substitution.) This approach, at the heart of present U.S. policy, sacrifices economic efficiency and competitiveness, violates market
principles and free-trade rules, and illogically supposes that the solution to domestic depletion is to deplete faster—a policy of “Strength
Through Exhaustion.” 90
• Trade, being nondiscriminatory, unsentimentally buys oil wherever it’s
cheapest. Such big economies as Japan and Germany import all their
oil, and pay for it by exporting goods and services in which they enjoy
a comparative advantage. U.S. oil imports are 24% of world oil trade,
yet major U.S. allies and trading partners import an even larger fraction of their oil, and many developing countries rely still more on
imports. Trade can be economically efficient, dominates world oil use,
sets its price, and underpins the global economy on which the U.S.
economy depends. But it exacerbates the issue of unreliable and vulnerable supplies, because for a fungible commodity (Box 2), everyone
shares shortages as well as surpluses.
• Substitution replaces oil with more efficient use or alternative supplies
wherever that’s cheaper. These domestic substitutes have all the
advantages of protectionism without its drawbacks, and all the advantages of trade without its vulnerabilities. Substitution offers a major
opportunity to all countries, both rich and poor. It has so far received
less attention and investment than it merits, but it generally turns out
to have lower costs and risks than buying oil in the world market, so
it forms this study’s main focus.
The value of substitution depends on the costs it incurs and avoids.
Besides business risks from insecure sources and brittle infrastructure,
America’s oil habit burdens us all with purchase price, foregone growth
and competitiveness, trade and budget deficits, inflated and volatile
prices, compromised public and environmental health, faster resource
depletion, damage to national reputation and influence, the prospect
of growing rivalry rather than friendship with countries like China
and India, and erosion of the global stability and security on which all
commerce depends. In 1975, 1979, 1981, 1989, 1995, and 2000, U.S. presidents indeed officially found that oil imports threaten to impair national
security.97 So what’s it worth to reduce U.S. oil dependence?
Oil Dependence
84. E.g., Peters 2002; PryceJones 2002; Frum & Perle
2003, p. 141; Barone 2002,
p. 49; Podhoretz 2002;
Henderson 2002; Murawiec
2002 (see also Ricks 2002
and Ricks 2002a); R.E. Ebel,
quoted in Dreyfuss 2003;
Margolis 2002; K. Adelman,
quoted in Marshall 2002; cf.
Buchanan 2003. The risks
today would probably be
far greater than envisaged
in CRS 1975, or by former
Ambassador Akins (Higgins
2004; see note 127).
85. Lovins & Lovins 1982.
86. Lovins & Lovins 2001,
particularly the detailed
annotations to the severely
abridged security discussion on p. 75. See also
online updates at
www.rmi.org, Library,
Energy, Security.
87. “The United States is a
high cost producer compared to most other countries because it has
already depleted its known
low cost reserves”:
DOC/BEA 1999.
88. The lower figure is from
the U.S. Energy Information
Administration, the higher
from the American
Petroleum Institute and
British Petroleum, both
using the canonical Oil and
Gas Journal denominator.
89. Foreign oil is also heavily subsidized (and its shipping lanes defended at taxpayer expense, pp. 20–21);
the leading independent
authority on energy subsidies (D. Koplow, personal
communication, 28 May
2004) says it’s uncertain
whether foreign or domestic oil is now more heavily
subsidized to U.S. users.
90. This phrase is due to
the late conservationist
and USA Tenth Mountain
Division Maj. (Ret.) D.R.
Brower.
91–96: See Box 2 on p. 14.
97. DOC/BEA 2001. These findings under the Trade Expansion Act of 1962 authorized the president to adjust oil imports, as was done in 1973 (licensing fees),
1975 and 1979 (tariffs), 1979 (Iran embargo), and 1982 (Libya embargo). In 1989, 1995, and 2000, the threat finding was still made, but no action beyond existing energy policy was deemed to be required.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
13
Oil Dependence
Oil supplies are becoming more concentrated and less secure: Domestic oil is limited
2: Oil is fungible
Many commentators imagine that we can simply shift from hostile to friendly oil sources. But
because oil is freely traded, the only path to
reliable supply and stable price is to use less
total oil, not just to import less oil. Although the
world market trades a great variety of crudes—
light and heavy, sweet and sour—crude oil is
basically a fungible (interchangeable) commodity.91 Saving 1 Mbbl/d therefore can’t reduce U.S.
imports from a particular country or region by
1 Mbbl/d: rather, world flows of oil will re-equilibrate at 1 Mbbl/d or ~1.3% lower supply and
demand (and at a slightly lower price92). If a lowcost producer like Saudi Arabia chose to undercut, say, costly marginal wells in the U.S., as has
happened before,93 then unconstrained imports
might not decrease at all. (The aim of the OPEC
cartel is to constrain supply, and thereby force
others to produce higher-cost oil first, then sell
the cartel’s cheap oil for that higher price—and
by depleting others’ oil first, make buyers even
more dependent on the cartel later.)
Cutting U.S. oil use by, say, one-eighth will thus
save the equivalent of imports from the Gulf (the
sort of comparison sometimes made in this
report to illustrate magnitudes), but one way or
another, the same amount of oil may still flow
from the Gulf to the U.S., as low-cost exporters
share reduced sales. Even the 1973 Arab
embargo couldn’t stop large oil flows from
reaching the U.S. by transshipment or by freeing
up other supplies. Saving oil is vital and valuable: it saves the purchase price, helps dampen
world prices, avoids indirect economic and
social costs, erodes the leverage of any particular supplier, and enables any given substitute to
meet a larger share of the decreased demand.
But if we keep the oil habit, we’ll still need a fix
from the same pushers in the same market.
The policy issue is the level of oil, not import,
dependence, so more “oil independence” is not
14
an optimal goal. As President George H.W.
Bush’s 1991 energy plan explained,94
Popular opinion aside, our vulnerability to price shocks
is not determined by how much oil we import. Our vulnerability to oil price shocks is more directly linked to:
(1) how oil dependent our economy is; (2) our capacity
for switching to alternative fuels; (3) reserve oil stocks
around the world; and (4) the spare worldwide oil production95 capacity that can be quickly brought on line.
And as President George W. Bush’s energy plan
96
rightly reiterated ten years later,
We should not…look at energy security in isolation
from the rest of the world. In a global energy marketplace, U.S. energy and economic security are directly
linked not only to our domestic and international energy supplies, but to those of our trading partners as
well. A significant disruption in world oil supplies
could adversely affect our economy and our ability to
promote key foreign and economic policy objectives,
regardless of the level of U.S. dependence on oil
imports.***The first step toward a sound international
energy policy is to use our own capability to produce,
process, and transport the energy resources we need
in an efficient and environmentally sustainable manner. Market solutions to limit the growth in our oil
imports would reduce oil consumption for our economy and increase our economic flexibility in responding to any…disruption. . . .
91. This is decreasingly true of U.S. gasoline because of a proliferation of “boutique” state and regional formulas. The resulting
difficulty of swapping supplies can create local shortages of gasoline meeting local regulatory requirements, even if there’s plenty
of total gasoline in the market.
92. Based on the Greene and Tishchishyna (2000) analysis, a 1%
decrease in U.S. oil demand reduces world oil price by ~0.3–0.5%;
p. 20 in NAS/NRC 2001 concurs.
93. For example, low oil prices from late 1997 to early 1999 squeezed
the ~7,000 independent operators who produce two-fifths of lower48 oil, forcing them to cut staff and exploration, increase debt,
and shut in or abandon high-cost wells. Such squeezes periodically
endanger nearly 1 Mbbl/d of marginal production, mainly from
small “stripper” wells.
94. DOE 1991, p. 3.
95. By industry and economic convention, we use the term “production” in this report, even though people only extract the oil that
geological processes have produced. We similarly use “consumption” in its economic sense; burned oil simply reacts with oxygen
to form combustion products of identical total weight, plus heat.
96. National Energy Policy Development Group 2001, pp. 8-3 and 8-1.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Counting the direct cost of oil dependence
Sending $2 trillion
abroad for oil and
empowering OPEC to
charge excessive
prices has cost the
U.S. many trillions of
dollars. Volatile oil
prices buffet the
economy, and still
would even if all our
oil were domestic—
to the extent we still
depend on oil.
In 2000, Americans paid $362 billion for retail oil. Although that’s $334 billion less than they would have spent at the 1975 oil/GDP ratio, they spent
nearly as much just on transportation fuel ($285 billion98) as on national
defense ($294 billion in FY200099). Transportation’s need for light refined
products, especially gasoline, causes ~93% of projected growth in oil
demand to 2025.100 Worse, it propels oil imports, which met 53% of U.S. oil
demand in 2000. Those imports’ net cost, $109 billion, was 24% of that
year’s goods-and-services trade deficit, now a major economic concern
that erodes the U.S. dollar’s value and hence boosts dollar oil prices.101
It’s getting worse. In 2003, imports cost $10 billion a month. And the bill
mounts up. During 1975–2003, Americans have sent $2.2 trillion (2000 $)
abroad for net oil imports. Those high and rising imports have strengthened OPEC’s ability to price oil above its fair market value. By sucking
purchasing power and reinvestment out of the U.S. economy, that pricing
power has incurred a total economic cost estimated at $4–14 trillion (1998
present-valued dollars) over the past three decades—about a GDP-year,
rivaling total payments on the
National Debt.102
1985
40
30
1979
1974
1991
1997
2003
10
1Q2004
(preliminary,
30 Apr 04
price)
2002
1998
1989
1999
85
80
75
70
1973
1970
65
0
2000
1987
20
60
101. Leone & Wasow (2004) correctly note that from the
end of February 2002 to two years later, the world oil price
rose 51% in dollars but only 4% in euros, as the exchange
rate plunged from 1.16 to 0.80 euros per dollar, for reasons
the authors ascribe largely to loss of international confidence in U.S. tax and fiscal policy. “In this situation, it is
perfectly rational for foreign suppliers of oil to charge
more in dollars to make up for the falling value of our currency.” On these figures, ~96% of the increase in the U.S.dollar oil price was simply an exchange-rate adjustment
relative to the euro, although OPEC spends dollars too.
1980
50
55
100. EIA 2004, p. 97.
1981
1983
50
99. DoD 2001, Table 1-11.
60
45
98. EIA 2003, Table A3, converting 2001 to 2000 $.
Figure 3: World oil consumption and real price, 1970–1Q2004103
Demand grew quickly at low prices until the 1973 “first oil shock,”
then slowly at high prices, until 1979’s worse “second oil shock” sent demand
into decline, softening price until demand began to rise again.
Prices spiked in the 1991 Gulf War. In 2003, the Iraq War and OPEC discipline
returned prices to 1974–79 levels, and in 2004, they rose further.
crude oil price
(1 Jan. Saudi 34º API light, 2000 $/bbl FOB)
In 2003, oil imports cost $10 billion a month.
During 1975–2003,
Americans sent $2.2 trillion (2000 $)
abroad for net oil imports.
In 2000,
Americans spent
nearly as much
on transportation fuel
as on national
defense.
102. Greene & Tishchishyna 2000. Competing, but in our
world oil consumption, annual average (Mbbl/d)
view unpersuasive, theoretical arguments are presented
by Bohi & Toman 1996. Kohl (2004) notes that the ORNL
Source: RMI analysis from EIA: March 2004 International Petroleum Monthly, “U.S. Petroleum Prices” 2004.
estimate assumes oil would compete at ~$10/bbl absent
the cartel, and that opinions differ on that assumption.
It is indeed controversial; yet noted oil economist Phil Verleger has reportedly estimated that real GDP would be higher by about 10% in the U.S., 15% in the
EU, and 20% in Japan if hydrocarbon prices had been determined by free markets over the past quarter-century.
103. Consumption data 1970–2003 from EIA 2004a, Table 46. Price data (applying implicit GDP price deflator) from EIA, undated; 2004 data are preliminary,
for 1Q only, and 30 April instead of 2 January data. Prices are opening-of-year snapshots, not annual averages, for the standard Saudi light marker crude,
excluding shipping costs.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
15
Oil Dependence
Counting the direct cost of oil dependence
3: The uncounted economic cost of oil-price volatility
Oil-price volatility (often measured by the standard deviation of log-price differences) incurs
up to 11 kinds of economic costs compared to
less- or nonvolatile alternatives. These costs
seem not to have been well explored in the economic literature, even though oil price volatility
is important, and might better explain output
and unemployment than the level of oil price
does.104 RMI commissioned a review of these
costs by researcher Jamie Fergusson, summarized in Technical Annex, Ch. 2. The uncounted
costs are partly overlapping, hard to assess
accurately, but large: one illustrative econometric analysis found on the order of $20b/y but
was methodologically unconvincing.105 At RMI’s
request, therefore, Dr. Richard Sandor, Vice
Chair of the Chicago Board of Trade, discovered
the volatility’s value in the derivatives market
using forward option pricing. As of 2 June 2004,
104. Hooker 1996; Awerbuch & Sauter 2002.
105. EIA 2001. The same is true of the ~$6–9 billion/y estimate by the
CRS (1992), updated by Moore, Behrens, & Blodgett (1997).
over a five-year time horizon, the market valued
oil-price volatility at ~$3.8/bbl, a ~10% premium.106 This figure, which we adopt as $3.5/bbl in
2000 $, would theoretically (not practically)107
equate to ~$7–8b/y to hedge the portion of U.S.
oil consumption traded in the market, or nearly
$30b/y if notionally extended to all current U.S.
oil consumption. (Volatility’s value is usually
higher over a shorter time horizon,108 and varies
over time according to real-time oil-price volatility itself: Technical Annex, Ch. 2) Displacing oil
with alternatives whose prices are less volatile
than oil’s, or may even be nonvolatile (such as
end-use efficiency, which is financially riskless109 because once installed it has zero price
fluctuation), would thus create major economic
benefits. Any risk premium on oil price must be
included in a properly risk-adjusted comparison
of oil vs. such less risky alternatives (p. 101).
The social cost of volatility may be higher than
that risk premium implies.
106. Dr. Sandor and his staff at the Chicago Climate Exchange (CCX) obtained 5-year “at-the-money” option pricing—prices for contracts to buy oil at
the 2 June 2004 price of $40, using the West Texas Intermediate (WTI) U.S. marker crude. CCE economist Murali Kanakasabai (personal communications, 26 May and 11 June 2004) used a $40 spot price, a strike price of $31.70 equal to the 5-y NYMEX futures price, and the 2 June 2004 forward price
volatility of 17% (from Lewis Nash, Morgan Stanley). Under these conditions, contracts on the New York Mercantile Exchange, valued using standard
Black-Scholes option calculators as of 2 June 2004 with a 5% riskless interest rate, were priced at $3.31/option for a call and $0.52 for a put for 5-year
Asian Options, which entitle the holder to buy oil for cash five years hence at the average price during the five-year period, thus separating secular
trends from mere fluctuations. The corresponding European Options cost $15.88 call and $0.572 put; American Options, $15.89 call and $0.774 put. (An
American Option costs more than the Asian Option because it entitles the holder to exercise anytime between purchase and expiration, rather than
only at expiration, exposing the option seller to the most extreme price of the underlying commodity during the entire period; a European option carries the same risk but can be exercised only on its expiry date.) We interpret the Asian Option call-plus-put price as a broad metric of volatility’s market value. (Banker’s Trust’s Tokyo office invented Asian Options in 1987 precisely for the purpose of laying off crude-oil price risk.) The owner of Asian
Options for both a call and a put has paid the sum of these instruments’ cost to eliminate price risk and be subject only to secular trend.
107. Extending the ~$4/bbl price discovered from small-quantity option quotations to the entire U.S. oil market (or at least the ~1.8 billion bbl/y traded in
the market—not the bulk of oil supply that comes from vertically integrated suppliers) is theoretical, in the sense that the market couldn’t stand the
hedging of such a large volume, nor is there a sufficiently large and creditworthy private counterparty. (Just the 5-year maturity would require a AAA
rating.) Despite this lack of a clear interpretation at large volumes, we believe the small-quantity marginal price is a helpful contribution to understanding the reality of oil-price volatility and its economic cost: as in any hedging transaction, the option seller is being compensated for a real risk.
108. This is because the market, correctly or not, assumes long-term reversion to mean. At 2 June 2004, exceptionally, the 1-year call-plus-put Asian
Options cost a total of $2.98/bbl, because the oil price was so anomalously high that most or all of the upside price risk was presumed to have already
occurred. However, the 90-day volatility price was higher than the 1-year, presumably because the market anticipated a greater short-term risk of
Middle Eastern supply disruptions. Going in the other direction, it would be possible to obtain longer-than-5-y option pricing, say for 7 or even 10 y,
but at that maturity, the market becomes quite thinly traded, and unfortunately there is no 20-y market deep enough to be validly quoted.
109. In financial economics, “riskless” means the price is known or constant, not that there are no other risks in a transaction (such as equipment
breakdowns or installer malfeasance). Thus a fixed-rate mortgage and the yield of a Treasury note are riskless. Such cashflows cost more because
they lack the price-fluctuation risk of an adjustable-rate mortgage or of the market value of a stock. (Treasury debt also yields less and costs more
than junk bonds because it lacks their credit risk.)
16
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Counting the direct cost of oil dependence
But the oil problem isn’t just about imports from unstable countries and
from a global cartel. That’s because replacing unstable suppliers with
diversified, stable, friendly sources is helpful but inadequate: even if the
U.S. imported no oil, it would still be a price-taker, its domestic markets
whipsawed by the world oil market’s price volatility (Fig. 3), which
exceeds that of the stock market or of most commodities. Oil price shocks
hurt far more than declines help, cutting the annual GDP growth rate
by about half a percentage point.110 Eight of ten postwar U.S. recessions
closely followed an oil-price spike,111 and according to Fed Chairman
Alan Greenspan, “All economic downturns in the United States since
1973…have been preceded by sharp increases in the price of oil.” 112 Each
$1/bbl price rise amounts to a gross tax of $7 billion per year (less any
return flows). For more than a century, oil prices have been random, just
like other commodity prices, increasingly exposing the U.S. economy—
as many key partners abroad have long been exposed—to costly shortand long-term perturbations. The market itself values oil-price volatility
at several dollars per barrel (Box 3).
Oil-price volatility, let alone cutoffs, poses major risks to key industries.
One American private-sector job in ten is linked to the auto industry,113 and
almost all jobs ultimately depend on mobility. The Big Three automakers
earn 60% of their global profits from North American sales of light trucks,
but surging oil prices in 2004 stalled those sales, swelling inventories and
requiring ~$4,000 incentives to “move the metal.” 114 Hard-pressed U.S. airlines, which lose $180 million a year for every penny-per-gallon rise in fuel
cost, may see their precarious gains scuttled.115 Independent truckers can’t
pay both their truck loans and sky-high fuel prices. As such bellwether
sectors falter, the damage reverberates through the whole economy. And
that’s just the tip of a giant iceberg of hidden security, fiscal, environmental, and foreign-relations costs.
Oil Dependence
“All economic downturns in the United
States since 1973…
have been preceded by
sharp increases
in the price of oil.”
— Alan Greenspan
104–109: See Box 3, p. 16.
110. This summary assessment by Hamilton is
described and supported in
Fergusson’s Ch. 2 of the
Technical Annex.
111. Kohl 2004.
112. Greenspan 2002.
113. Including multipliers
from the direct employment
of 1.3 million direct employees of automakers and 2.2
million employees of the
industry’s suppliers, plus the
rest of the value chain
(McAlinden, Hill, & Swiecki
2003). The multiplier is large
because the motor-vehicle
manufacturing sector adds
value of $292,000/worker,
143% above the U.S. manufacturing average—the thirdhighest value-added per
worker of any major sector,
with wages also 60% above
the average U.S. worker’s.
The total reaches 13.3 million
if it also includes 3.9 million
professional drivers.
114. Freeman, Zuckerman, &
White 2004.
115. Trottman 2004.
Oil dependence’s hidden costs
may well exceed its direct costs
One-fourth of all the oil America uses comes from OPEC countries,
one-seventh from Arab OPEC countries, one-eighth from Persian Gulf
countries. Reliance on unstable oil sources incurs costs for both buying
and defending it. Those costs are compounded by how some oil exporters
respend the petrodollars. The $2.2 trillion paid for U.S. oil imports since
1975 has financed needed development by neighbors and allies, but has
also paid for profligacy, polarizing inequities, weapons of mass destruction, state-sponsored violence, and terrorism—perhaps indirectly and
unofficially including the 9/11 attack on the American homeland.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Military, subsidy,
and environmental
costs could at least
double oil’s true price
to society.
One-fourth of all the oil
America uses comes
from OPEC countries,
one-seventh from
Arab OPEC countries,
one-eighth from
Persian Gulf countries.
17
Oil Dependence
Oil dependence’s hidden costs may well exceed its direct costs
Oil dependence
creates incentives
and structures that
make most oil-exporting countries
uniquely ill-governed,
unfree, unjust,
unstable, militarized,
and conflict-prone.
Petrodollars tend to destabilize
116. The “Publish What You
Pay” campaign (www.publishwhatyoupay.org) nearly
got BP expelled from
Angola in 2001, but is now
becoming the “gold standard” for extractive industries, especially in Africa.
The World Bank Group
endorsed the Extractive
Industries Transparency
Initiative in 2003 (www.
worldbank.org/ogmc/) and
is considering in 2004 the
wider and more controversial recommendations
of the Extractive Industries
Review, which include phasing out oil-project investments within five years.
116a. Moody-Stuart 2003.
117. DoD sought funding in
2003 to train Colombian soldiers to protect assets of
private U.S. oil operators
there (BBC News World
Edition 2003). The two main
pipelines, each over 400
miles long, suffered 98
bombings in 2000 alone
(Karl 2004).
118. Klare (2004) gives
details.
119. Klare (2001) lists (pp.
227–231) 19 separate conflicts then underway over
oil, involving an overlapping
total of 49 countries.
120. Karl 2004.
121. Luft & Korin (2003),
citing Freedom House 2003
and Transparency International 2003.
122. Gary & Karl 2003, pp.
77, 18; see also Karl 1997.
The $200 billion, chiefly
from the Gulf of Guinea, is
about ten times expected
Western aid flows.
123. Nunn et al. 2000.
18
The costs of oil imports (broadly defined) include political ties to unstable
countries. Prospects of those countries’ achieving greater political stability
simply by acquiring oil wealth are slim, as historic experience amply
demonstrates. There are some commendable exceptions, and recent initiatives by BP and others to post on the Web the amounts and recipients of
all payments to governments116 are important and laudable blows against
corruption. Oil-driven development failure is not inevitable.116a But more
often, oil wealth has fomented power struggles among governmental,
sociopolitical, and industrial factions. Such rivalries, social divisions, and
developmental distortions can threaten price and supply stability, incur
escalating protection costs,117 and overstress oil reservoirs (as under
Saddam Hussein). Indeed, countries often become unstable once they discover oil. Of the top eight non-Gulf oil prospects—Angola, Azerbaijan,
Colombia, Kazhakstan, Mexico, Nigeria, Russia, and Venezuela—not one
is stable.118 In some, buying rights to a new multi-billion-dollar oilfield
might mean bargaining with both the government and other sellers such
as rebels, clan chieftains, warlords, or druglords.119
Flows of oil revenue to governments that start off neither transparent nor
fair tend to create incentives and structures that then exacerbate destabilizing corruption, inequity, and repression:120 only 9% of world oil reserves
are held by countries considered “free” by Freedom House, and oil riches
correlate well with Transparency International’s corruption ratings.121
An NGO report on the more than $200 billion destined for sub-Saharan
African governments in the next decade from their rapidly expanding oil
reserves notes:122
The dramatic development failures that have characterized most other oildependent countries around the world…warn that petrodollars have not helped
developing countries to reduce poverty; in many cases, they have actually exacerbated it.***Countries that depend upon oil exports, over time, are among the
most economically troubled, the most authoritarian, and the most conflict-ridden
states in the world today.
Nigeria, for example, has received more than $300 billion in oil revenues
in the past quarter-century, but its per-capita income remains below $1 a
day, and its economy has performed worse than that of sub-Saharan
Africa as a whole, let alone other developing regions. The Center for
Strategic and International Studies’ energy task force found in 2000123 that
the nations now counted upon to moderate U.S. Gulf dependence often
…share the characteristics of “petro-states,” whereby their extreme dependence
on income from energy exports distorts their political and economic institutions,
centralizes wealth in the hands of the state, and makes each country’s leaders
less resilient in dealing with change but provides them with sufficient resources
to hope to stave off necessary reforms indefinitely.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil dependence’s hidden costs may well exceed its direct costs: Petrodollars tend to destabilize
The vicious circle of rent-seeking behavior and concentrated power also
often reinforces prevalent over-militarization, which in turn is both internally and externally destabilizing:124
In the decade from 1984 to 1994…OPEC members’ share of annual military
expenditures as a percentage of total central government expenditures was three
times as much as the developed countries, and two to ten times that of the nonoil developing countries. From the perspective of poverty alleviation, the sheer
waste of this military spending is staggering….Fights over oil revenues become
the reason for ratcheting up the level of pre-existing conflict in a society, and oil
may even become the very rationale for starting wars. This is especially true as
economies move into decline. Petroleum revenues are also a central mechanism
for prolonging violent conflict and only rarely a catalyst for resolution. Think,
for example, of Sudan, Algeria, the Republic of Congo, Indonesia (Aceh),
Nigeria, Iraq, Chechnya [a key transit point for Caspian oil pipelines] and
Yemen.
Of these eight states, at least seven, not coincidentally, now harbor
Islamic extremists.125
Sociopolitical instability drives military costs
Thus arise threats to regional stability: the wars, terrorist movements,
and decaying state structures that the U.S. Central Command’s forces must
address in order to keep the oil flowing. But that mission presupposes that
the United States needs the oil (or is willing to protect it for others’ sake).
Financially, net of other countries’ ~$54 billion contributions, the 1991 Gulf
War cost the U.S. only $7 billion—equivalent to a $1/bbl price increase for
a year. Yet spending $7 billion each year to buy oil efficiency technologies
then available could have permanently eliminated the equivalent of Gulf oil
imports.126 This comparison suggests a serious and continuing misallocation of America’s capital and attention.
The United States has multiple interests in the Middle East, including its
commitment to Suez Canal/Red Sea navigation and to Israel. But those
interests are at best distorted and at worst overwhelmed by the vital need
for oil. Three decades ago, President Nixon was even prepared as a last
resort to use airborne troops to seize oilfields in Saudi Arabia, Kuwait,
and Abu Dhabi—perhaps for up to a decade—if the Arab oil embargo
weren’t lifted, because the U.S. “could not tolerate [being]…at the mercy
of a small group of unreasonable countries.” 127 Historians will long debate
whether the United States would have sent a half-million troops to liberate Kuwait in 1991 if Kuwait just grew broccoli and the U.S. didn’t need
it. Decades hence, historians may be better able to say whether an odious
tyrant would have been overthrown with such alacrity in 2003 if he didn’t
control the world’s second-largest oil reserves. But even in peacetime,
without lives directly at stake, the United States has for decades routinely
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Countries often
become unstable
once they discover oil.
Financially, the 1991
Gulf War cost the
U.S. only $7 billion;
spending less than this
to buy oil efficiency
instead could have
eliminated the equivalent of Gulf oil imports.
Dependence on oil
from the world’s
least stable region
has created the
largest single burden
on America’s military
forces, their costs,
and their likelihood
of success.
124. Gary & Karl 2003, p. 24.
125. Their presence is obvious in all but the Republic
of Congo, whose weak
state structure nonetheless
sustains activities that
would be useful support
structures for terrorism,
such as money-laundering
and illicit arms sales.
126. Lovins & Lovins 2001,
hypertext to p. 77, “cost
the United States.” Today’s
technologies are even better
(Fig. 28, p. 100 below), and
the cheapest ones have
negative or very low costs.
127. Frankel 2004; Alvarez
2004. According to Higgins
(2004), James Akins, the
U.S. Ambassador to Saudi
Arabia, sent a confidential
cable to Washington calling
the notion of seizing Saudi
oil “criminally insane,”
and was fired a few months
later—because of that view,
he says.
19
Oil Dependence
Oil dependence’s hidden costs may well exceed its direct costs: Sociopolitical instability drives military costs
We pay two to three
times as much to
maintain military
forces poised to intervene in the Gulf as
we pay to buy oil from
the Gulf.
maintained large forces whose stated primary mission is intervention in
the Persian Gulf. In every issue of their United States Military Posture
Statement from FY1979 to FY1989, the Joint Chiefs said this was even more
important than containing Soviet ambitions in the region.128 The 2002
U.S. national-security strategy mentioned neither oil nor the Gulf,129 but
the 18 February 1992 draft of Defense Planning Guidance for the Fiscal Years
1994–1999 forthrightly stated America’s “overall objective” in the Gulf:
“to remain the predominant outside power in the region and preserve
U.S. and Western access to the region’s oil.”130 Big U.S. military bases now
blanket the area. By 2000, prominent conservatives wrote,131
128. Delucchi & Murphy
1996; extensive further
evidence is marshaled
by Klare (2004).
The presence of American forces, along with British and French units, has
become a semipermanent fact of life....In the decade since the end of the Cold
War, the Persian Gulf and the surrounding region has witnessed a geometric
increase in the presence of U.S. armed forces, peaking above 500,000 troops during Operation Desert Storm, but rarely falling below 20,000 in the intervening
years....[R]etaining forward-based forces in the region would still be an essential element in U.S. security strategy given the longstanding American interests
in the region***[where] enduring U.S. security interests argue forcefully for an
enduring American military presence.
129. The White House 2002.
130. Quoted, with extensive
historical analysis, in
Delucchi & Murphy 1996.
131. Donnelly, Kagan, &
Schmitt 2000, pp. 14, 17, 74.
132. Koplow & Martin 1998,
Ch. 4; ORNL 2003, Table 1.9.
The estimates are by Earl
Ravenal (Cato Institute and
Georgetown U. School of
Foreign Service), William
Kaufmann (Brookings
Institution), and Milt
Copulos (National Defense
Council Foundation), whose
detailed unit-by-unit cost
assessment included no
general DoD overheads, yet
yielded peacetime readiness costs of $49.1 billion
(2002 $).
Gulf-centric forces’ 1990s readiness cost, allocating DoD’s non-regionallyspecific costs pro rata, was ~$54–86 billion per year in 2000 $.132 That is,
the U.S. pays two to three times as much to maintain military forces poised to
intervene in the Gulf as it pays to buy oil from the Gulf. By claiming that we’d
need most of those forces anyway, and that their presence is largely altruistic, some analysts select numerators and denominators that can translate
$54–86b/y into per-barrel costs around $2 rather than, say, $77.133 Econometrics134 and calculations of military-plus-Strategic-Petroleum-Reserve
costs135 estimated just $10/bbl before 1991. But all these figures understate
the distortion of paying oil-related military costs in taxes and blood, not
at the pump,136 because they count only Central Command. The Pentagon’s
other unified commands are also having to set aside many of the other
missions that they have trained for, as they are “slowly but surely being
133. Koplow & Martin (1998) very conservatively allocated only one-third of this military cost to oil-related missions, then divided it by all oil exports from
the Gulf (mainly to Europe and Japan, which get about one-fourth and three-fourths of their oil from the Gulf, respectively, but contribute inversely to the
Gulf’s G7 military costs: Johnson 2000, p. 87). Accepting the rationale that U.S. consumers benefit too from resulting improvements in price stability and
global economic health, the U.S. military subsidy cut the cost of Gulf exports to all countries by ~$1.65–3.65/bbl, or 4–9¢/gal. But one might wonder whether
an oil-in dependent United States would be quite so altruistic. If not, a different cost allocation could be warranted. If the entire nominal $70 billion of 2000
Gulf-related military costs were allocated to the one-eighth of Gulf exports that the U.S. imports, they’d be equivalent to $77/bbl—2.7 times the landed price
of Saudi crude in 2000—and would boost the 2000 pump price to $4.29 a gallon. The complexities of assessing the military costs of oil dependence are well
discussed by Delucchi & Murphy (1996).
134. Broadman & Hogan 1988.
135. Hall 2004.
136. The belief that this distortion is large spans a wide spectrum of politics and methodology. For example, M.R. Copulos (2003) reckons that imported oil
incurs U.S. economic costs of $5.28/gal of gasoline, annually comprising (2002 $) $49 billion military cost, $99 billion direct import cost, $61 billion indirect cost
of sending that money abroad, $13 billion lost tax revenue on the wages of domestic petroleum workers assumed to supply the oil instead, and $75–83 billion
from amortizing three historic oil shocks’ economic cost (totaling $2.2–2.5 trillion) over the 30 years they spanned.
20
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil dependence’s hidden costs may well exceed its direct costs: Sociopolitical instability drives military costs
converted into a global oil-protection service”137—for example, Southern
Command in Colombia, European Command in Africa (except the Horn)
and the Republic of Georgia, Pacific Command in two oceans, and
Northern Command in undisclosed places. DoD doesn’t do activity-based
costing, so how much of its peacetime budget relates to oil is unknown.
But if you think, for example, that half might be a reasonable estimate,
and should be ascribed to oil use by the United States but not by its oilusing allies, that would be equivalent to ~$25/bbl.
Nonmilitary societal costs
Oil dependence also bears other important hidden costs to U.S. society:
• Direct federal subsidies to both domestic and imported oil in 1995
added several dollars per barrel, may well have been higher for
imports (disadvantaging domestic producers), and are rising.138
Oil Dependence
Subsidies and
environmental costs
(including climate
risks), combined with
oil’s military costs,
could easily cost
society more than the
oil price we pay in
the market.
• Federal net subsidies to oil-using systems are probably larger still—
by one reckoning, $111 billion a year just for light vehicles,139
equivalent to $16/bbl.
• Oil’s unmonetized air-pollution costs have been estimated at
~$1/bbl140 to ~$15/bbl.141
• That doesn’t yet count ~$2–5/bbl for oil combustion’s potential
contribution to climate change.142 That may well be understated for
irreversible processes that threaten to submerge some countries, starve
or flood others, and export environmental refugees to others. This
could in turn destabilize whole regions, raise major national-security
contingencies with possibly tremendous associated costs,143 and
increase catastrophic-risk costs to the private sector.144
Whatever all these
hidden costs turn out
to total—clearly
upwards of $10/bbl,
plausibly comparable
to oil’s market price,
and perhaps much
higher yet—they
should hide no longer.
137. Klare 2004, p. 7; Cummins 2004a (~30 U.S. warships now patrol in and around the Persian Gulf); Glanz 2004.
138. Subsidies are complex, arcane, and often artfully concealed (e.g., by waiving normally required payments or rules), but their net effect is to transfer
government-provided goods, services, or risk-bearing to private firms that must otherwise buy them in the marketplace. In 1995, annual nonmilitary subsidies
to the U.S. petroleum industries (both domestic and via imported oil, net of federal revenues from oil-industry user fees and consumer excise taxes) totaled
upwards of $5.2–11.9 billion; those to domestic oil, $4.4–10.2 billion, totaled $1.2–2.8/bbl according to Koplow & Martin (1998; see also Koplow 2004). Their
analysis included subsidies mainly in tax breaks, R&D support, subsidized credit, below-market resource sales, subsidized oil transport, and socialization of
private-sector liabilities, but excluded subsidies to the car, aircraft, and other oil-using industries; subsidies from eight hard-to-analyze federal programs;
effects on oil supply and demand and on employment; and leveraging of private investment into the oil sector beyond the levels attractable at fair market
prices. Some other oil-subsidy estimates are much higher. For example, ICTA (1998) estimate annual subsidies to U.S. oil (1997 $, including defense) at
$126–273 billion plus $0.4–1.4 trillion in externalities of using the oil, including all side effects of the transportation system. (For example, vehicular combustion products can be a major threat to public health, especially in the teeming cities of the developing world. Congestion, collision, and inequitable access to
mobility add to the social toll.) ICTA therefore finds an order-of-magnitude underpricing of gasoline, because of externalities totaling $4.6–14.1/gal.
Conversely, the American Petroleum Institute strongly disputes Koplow & Martin 1998 (Dougher 1999); they have responded
(www.earthtrack.net/earthtrack/index.asp?pageID=144&catID=66).
139. Roodman 1998, Ch. 5.
140. NAS/NRC (2001, Ch. 5, summarized at p. 86), estimated gasoline’s environmental externalities at ~14¢ per U.S. gallon (~$35 billion/y in 2000), or $6 per
barrel, with a range from one-fifth to twice that much “not implausible” and perhaps not inclusive; 12¢ of the 14¢ was for climate change, which we show
separately in the next bullet.
141. Hall 2004 and that article’s journal citations.
142. NAS/NRC 2002a and Hall 2004, respectively.
143. Stipp 2004, Schwartz & Randall 2003.
144. See next page.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
21
Oil Dependence
Oil dependence’s hidden costs may well exceed its direct costs: Nonmilitary societal costs
Whatever all these hidden costs turn out to total—clearly upwards of
$10/bbl, plausibly several tens of dollars per barrel, and perhaps even
higher—they should hide no longer. President Nixon got the economic
principle of truthful pricing right when he told the Congress in 1971,145
Oil dependence
compromises fundamental U.S. security
and foreign-policy
interests and ideals,
harms global
development,
and weakens the
nation’s economy.
And even if world oil
output peaks later
than some fear,
it’s none too early to
begin an orderly
transition that could
capture its manifest
profits sooner.
U.S. oil dependence
is intensifying competition over oil with
all other countries,
ultimately including
China and India.
It sets the stage for
billions of people to
blame their poverty
and oil shortages on
what demagogues
could portray as
America’s uncaring
gluttony.
One reason we use energy so lavishly today is that the price of energy does not
include all the social costs of producing it. The costs incurred in protecting the
environment and the health and safety of workers, for example, are part of the
real costs of producing energy—but they are not now all included in the price of
the product.
Adding up the hidden costs
To be sure, despite all of oil’s direct and hidden costs, oil revenues are
important drivers of national development in such places as Mexico,
Russia, and potentially West Africa, the Middle East, and the Caspian
region. The need to diversify economies trapped in petro-codependency,
the needs of oil-dependent industries, and oil’s pivotal role in today’s
economy must be dispassionately balanced against the benefits of oil displacement. But taken together, the costs of continuing America’s current
dependence on oil pose a potentially grave impediment to achieving stated national security and economic objectives. In sum, U.S. oil dependence:
• erodes U.S. national security by:
• engaging vital national interests in far-off and unfamiliar places
where intervention causes entanglement in ancient feuds and
grievances, and even in oil wars;
• requiring military postures—such as deployments in the midst of
proud traditional societies—that reinforce Islamist arguments and
Islam/West friction, arousing resentment and inciting violence
among some of the world’s 1.3 billion Muslims; 146
• thereby turning American citizens and assets worldwide into
symbolic targets;
• providing resources for legal but excessive and destabilizing arms
acquisitions: as a 2000 CIA assessment dryly remarked, OPEC
revenues “are not likely to be directed primarily toward core
human resources and social needs”147 despite staggering development deficits;148
144. As “global weirding” (in
Hunter Lovins’s phrase) shifts
weather patterns and increases
weather’s volatility, the escalating costs of drought, fires, ocean temperature changes, devastating storms, crop and fishery failures, etc. could double catastrophic losses to $150 billion in
ten years, including insured losses of $30–40b/y (Heck 2004). U.S. oil subsidies are equivalent to ~$7–59 per ton of carbon emitted, so they more than offset
the value of CO2 abatement discovered in emerging carbon markets (Koplow & Martin 1998).
145. Nixon 1971.
146. Johnson 2004.
147. NIC 2000, p. 5.
148. Most of all in the 280-million-person,
22-nation Arab world (UNDP 2002/3; Economist 2002).
22
If you worry that depletion may happen sooner rather than later,
this study offers profitable and practical solutions;
if you’re a depletion skeptic, it’s a negative-premium insurance policy
against the chance that the depletionists might prove right.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil dependence’s hidden costs may well exceed its direct costs: Adding up the hidden costs
• giving oil-exporting states extra leverage to demand to be sold
advanced weapons that may later be turned against the U.S.
or its allies;148a
• providing resources for state-sponsored terrorism and for rogue
states’ development and fielding of weapons of mass destruction; 149
• requiring an unusually centralized energy production and transportation system inherently vulnerable to terrorism;
• constrains U.S. actions, principles, ideals, and diplomatic effectiveness by:
• distorting relationships with,150 and appearing to apply double
standards in dealing with, oil-producing states;
• undermining the nation’s moral authority by making every issue
appear to be “about oil” and national policy in thrall to oil interests: this is arguably one of the most important contributors to
rampant anti-American sentiment in much of the world—hostility
that has itself “become a central national security concern”;151
• accelerating the militarization of foreign policy at the expense of
the international norms, institutions, and relationships crafted by a
century of diplomacy;152
• injecting climate-driven irritants into relations with current partners,
such as Europe and Japan, and potential ones, such as China and
India, whose long-term friendship is a key to robust counter-terrorism collaborations and many other elements of global stability;
• intensifying competition over oil with all other countries, ultimately
including China and India—a likely path not to friendly relationships but to geopolitical rivalries akin to those that helped to trigger World War II;153
• setting the stage for billions of people to blame their poverty and
oil shortages on what demagogues could portray as America’s
uncaring gluttony;
• retards global development, perpetuating injustice and breeding
unrest, by:
• supporting unaccountable governments and undiversified
economies based on resource extraction at the expense of balanced,
broad-based development;
• thereby perpetuating regimes unable to meet the growing aspirations and demands of their populations, thus heightening political
tension, instability, and extremism;
• engendering corruption, opacity, rent-seeking, and concentration
of wealth, all of which can be exploited by terrorists and criminal
cartels, including the drug trade, and can undermine emerging
democratic institutions and norms of human dignity;
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
148a. Roberts (2004a) reports
Chinese offers of ballistic
missiles for Saudi oil.
149. Outside North Korea,
most proliferation of
weapons of mass destruction appears to be financed
by oil revenues and motivated by oil-linked rivalries;
current concerns focus on
oil-funded terrorists’
becoming a customer for
North Korean bombs.
However, oil funding is
abundantly sufficient but
not necessary as a revenue
source for terrorists and
proliferators: they could still
find other funds, e.g. from
the drug trade, even if their
oil-derived revenues fell to
zero. A world buying no oil
may well sprout less terrorism, but more by trimming
its roots than by cutting off
its water.
150. Partly through unilateralist foreign policy, in the
view of Daalder & Lindsay
(2003). But probably all foreign-policy experts would
agree that, as DOE’s 1987
report to the president,
Energy Security, put it,
“Increased dependence on
insecure foreign oil supplies reduces flexibility in
the conduct of U.S. foreign
policy.”
151. Today, “[T]he bottom
has indeed fallen out of
support for the United
States,” and “hostility
toward the United States
makes achieving our policy
goals far more difficult”:
Djerejian 2004.
152. Priest 2003.
153. From Iraq to the sixclaimant Spratly (aka
Spratley) Islands, from the
Baku subplot of World War
I to the Ploesti and East
Asian oil roots of World
War II to the oil-driven
intrigues and tyrannies and
civil wars of West Africa
and Central Asia, oil’s
wasteful use and concentrated supply remain preeminent among causes of
rivalry and conflict.
23
Oil Dependence
Oil dependence’s hidden costs may well exceed its direct costs: Adding up the hidden costs
•
4: Hedging
the risk
of oil
depletion
retards global development (continued)
• increasing oil prices and hence the unserviceable Third World
debt:154 in 2001, low- and middle-income countries’ fuel (mainly oil)
imports equaled two-thirds of their new borrowings; 155 and
• weakens the national economy by:
Former Shell
USA chief geologist M. King
Hubbert predicted in 1956
that U.S. oil production would
peak in the
early 1970s and
then decline. It
peaked in 1970.
In 1974 he predicted world oil
production
would peak in
1995. His successors, such
as noted petroleum geologists
Colin Campbell
and Kenneth
Deffeyes, now
predict that
peak around
2004–2010.
They warn that
the resulting
shift from a
buyer’s to a
seller’s market
would trigger
higher prices
and ultimately
severe economic disruption if
• imposing huge deficit-financed burdens on the U.S. for military
forces able to protect and secure access to oil and to deter mischief
in oil-related regions;156
• increasing U.S. trade deficits, compromising currency reserves and
the strength of the U.S. dollar, thereby straining import-dependent
industries and monetary policy while encouraging exporters to
raise dollar oil prices, or even redenominate oil trade in euros, and
to shift their reinvestments into stronger currencies;
• escalating price volatility and supply risks that hazard not only
oil-sensitive sectors, such as automotive and aviation, but ultimately the entire economy;
• extracting from unduly influenced legislators ever larger deficitfinanced domestic oil subsidies (which distort markets by suppressing fair competition and retarding cheaper options that could
reduce national costs);
• creating major environmental liabilities both at home and abroad,
increasing social and economic pressures, raising health-care costs
and lost labor productivity, adding expense to development projects and reclamation efforts, and raising the risk of costly and
irrevocable climate change;
• incurring large domestic opportunity costs in jobs, education,
environment, and other benefits achievable through a different
allocation of national resources.
Thus even if nothing goes badly wrong in the Middle East, the routine
direct and indirect costs of oil dependence, plus the rising contingent
risks of serious political or terrorist disruptions of supply elsewhere, are
compelling arguments for using less oil. They would remain so even if
the ultimate depletion of affordable global oil resources were infinitely
remote. But in fact the industry’s only debate about economic depletion is
timing (Box 4), and all experts predict a worrisome and inexorable rise in
Gulf dependence regardless of timing.
154. The debt crisis is rooted in the dizzying 150% increase in the foreign debt of 100 developing countries during 1973–77:
IMF 2002.
155. Imports of ~$147b (World Bank 2003, Table 4.6) vs. disbursements of new debt (World Bank 2003a). B. Bosquet (World
Bank), personal communication, 21 January 2003.
(continued next page)
156. For the broader resource context of such issues, see Klare 2001.
157–160. See Box 4.
24
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil dependence’s hidden costs may well exceed its direct costs: Adding up the hidden costs
4: Hedging the risk of oil depletion (cont.)
policies weren’t adjusted timely and we noticed
the shift too late to respond gracefully.157 In some
circles, these predictions arouse deep anxiety.158
The depletion debate159 is partly about geology
and the integrity of and agendas behind various
parties’ published oil-reserve and -resource
data; most key data are proprietary and opaque,
many are disputed, “geological” reserves differ
from “political” reserves (on which OPEC quotas
depend), and many producers cheat, leaving
tanker-counting scouts to estimate actual shipments. Differing units and definitions,160 reporting
cumulative discoveries as if they were remaining reserves, and not backdating reserve declarations to discovery dates, all add to the confusion. (To an economic geologist, “reserves” are
resources profitably exploitable with present
technology, hence varying with price and time
even without new discoveries. Economic theorists, for whom reserves are simply an inventory
replenished by investment, presume supply
forecasts’ historic understatement will continue.
Geologists are increasingly split.)
Unconventional oil, such as newly competitive
(though capital-, water-, and energy-intensive)
Albertan tar sands, may or may not be included.
157. Sources on this view are collected at www.peakoil.net and
www.peakoil.org. See also Bakhtiari 2004, pp. 19–21, and Berenson 2004.
158. For example, British Environment Minister (1997–2003) Michael
Meacher (Meacher 2004) warns that “if we do not immediately plan to
make the switch to renewable energy—faster, and backed by far greater
investment than currently envisaged—then civilisation faces the sharpest
and perhaps most violent dislocation in recent history.”
159. Laherrere (2004) provides an unusually clear summary of all the key
supply-side issues. Greene, Hopson, & Li (2003) provide a more optimistic
non-geological overview, summarized in Greene, Hopson, & Li 2004. Yet
they still find non-Middle-East oil production is likely to peak by 2025. The
timing of the Middle East production peak is less certain because the key
geological and reservoir-engineering data are secret and disputed.
160. For example, in 2003, differences between SEC rules (which require
an accounting standard of proof) and normal industry practice based
on geologically probable reserves, as well as internal assessment and
reporting problems linked to decentralized management, led Shell to
debook 3 billion bbl of petroleum reserves.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
And should past demand growth be extrapolated, or will substitutions automatically avert
scarcity? Most industry strategists say that
while oil resources are finite, modern technologies for finding and extracting them at declining
real cost are so powerful that barring major
political disruptions—a key and sanguine
assumption—oil depletion is unlikely to cause
big problems for decades. Such analysts reject
depletionists’ concern of physical shortages
unpleasantly soon. We agree. But historically,
major shifts in energy supply have also required
decades, so if world production will peak in (say)
2020–40, a common view among oil optimists,
then starting action now would seem sensible.
If you worry that depletion may happen sooner
rather than later, this study offers profitable and
practical solutions; if you’re a depletion skeptic,
it’s a negative-premium insurance policy against
the chance that the depletionists might prove
right. Such hedging seems prudent for three
basic reasons. Peak production, whenever it
occurs, will be visible only in hindsight, too late
to respond. Both producers and consumers have
strong market disincentives to express any concern about depletion, so farsighted public policies lack wide support. And the market inversion
caused by cartel behavior—OPEC’s output
restrictions force others to extract costlier oil
first—masks the smooth price run-up that would
otherwise provide early warning of depletion.
EIA, among the most optimistic resource estimators, thinks world oil output will peak sometime
after 2020 or 2030—but it also projects rapidly
rising oil demand in that period, implying a collision. The transitional dynamics analyzed in this
study therefore suggest that it’s none too soon to
begin graceful adaptation—especially if postponing the displacement of oil delays major
profits by needlessly prolonging the use of oil
that costs more than its substitutes.
25
Oil Dependence
If, as we’ll show,
oil’s replacements
work better and
cost less, then buying
them instead would
make sense and make
money. This would be
true even if oil had no
hidden costs to society.
To the extent oil’s
replacements proved
more lucrative than
oil itself, even for oil
companies,
the option of
dramatically reducing
oil dependence
would become
an imperative.
Contrary to some
economic theorists’
presumption that all
efficiency worth
buying has already
been bought, plenty
of low-hanging fruit
has fallen down
and is in danger of
mushing up around
the ankles. Noticing
and picking it offers
enticing business
opportunities.
161. Bizarrely, this is sometimes assumed even for
mixed and nonmarket
economies.
26
Could less oil dependence be
not only worthy but also profitable?
In 1987–88, this report’s senior author was asked by Royal Dutch/Shell
Group Planning to estimate how much of the oil used by the United
States in 1986 could have been saved if the most efficient oil-using technologies demonstrated in 1986 had been fully deployed in that year, and
if similar savings in natural gas had been substituted for oil in the furnaces and boilers where they’re interchangeable. The shocking conclusion
(Technical Annex, Ch. 3): 80% of 1986 U.S. oil use could have been saved at
an average 1986 cost of $2.5/bbl. This was said to be “very influential” in
shaping Shell’s thinking about how supplying more oil would have to
compete with its more efficient use; where an energy company could find
profits on the demand side; and what market risks its supply-side investments might face if customers bought efficient use instead.
The potential for oil saving and gas substitution, when far more closely
examined, remains impressive 18 years later: the technologies have
improved by even more than their potential has been used up. Moreover,
other supply-side oil displacements that didn’t look so attractive 15 years
ago—by biofuels and waste-derived fuels, even by hydrogen—are now
possible and potentially profitable. This study therefore explores for the
first time the combined potential of all four ways to displace oil by domestic (or North American) resources, and hence to surmount the nationalsecurity, economic, and environmental challenges of oil dependence. And
it suggests that much if not all of this diverse portfolio of opportunities
for the next energy era may also enable oil companies and oil-exporting
countries to make more profit at less risk than they do now (pp. 248–257).
To the extent this were true—to the extent oil’s replacements proved better,
cheaper, and more lucrative than oil itself, even for oil companies—the
option of dramatically reducing oil dependence would become an imperative. Using less oil would not only reduce or ultimately eliminate the
many risks just described; it would make money, even in the short term.
But if that bonanza were real, why wouldn’t it already have been captured?
Beliefs that hold us back
A central dogma of dimly recalled Economics 101 courses, widely sloganized by pundits (good economists know better), holds that existing markets are essentially perfect,161 so if smarter technologies could save energy
more cheaply than buying it, they’d already have been adopted. Actually,
perfect markets are only a simplifying assumption to render theory tractable. Practitioners of energy efficiency find it utterly contrary to their
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Could less oil dependence be not only worthy but also profitable? Beliefs that hold us back
experience. Every day they battle scores of “market failures”—perverse
rules, habits, and perceptions that leave lucrative energy savings
unbought. A huge literature documents these obstacles and explains how
to turn them into business opportunities.162 Armed with such “prospector’s guides,” firms like DuPont,163 IBM, and STMicroelectronics are cutting their energy use per unit of output by 6% every year with 1–3-year
paybacks. That’s among the highest and safest returns in the whole economy. If markets were perfect—if it’s not worth picking up that $20 bill
lying on the ground, because if it were real, someone would have picked
it up already—these and other real-world examples of big, cheap, rapid,
continuing energy savings couldn’t exist.164 In fact, if markets were perfect, all profit opportunities would already have been arbitraged out and
nobody could earn more than routine profits.165 Fortunately, that’s not
true: the genius of the free-enterprise system lets entrepreneurs create
wealth by exploiting opportunities others haven’t yet noticed. This messy,
squirming hubbub of innovation, even to the point of “creative destruction,” is at the heart of wealth creation.
The old debate about how much energy can be saved at what cost by
more productive technologies has largely ascended to the rarified plane of
theology, partly because the policy priesthood includes more ordinary
economists than extraordinary engineers, and the two rarely connect.
We therefore alert you to this dispute up front. If you believe that economic theory is reality and physical measurement is hypothesis, you may
doubt many of our findings.166 Market competition will ultimately reveal
who was right. But if you’re a pragmatist who weighs the evidence of
costs and savings rigorously measured by engineers in actual factories,
buildings, and vehicles; if you believe that price is an important but not
the sole influence on human behavior; if you consider economics a tool,
not a dogma; and if you think markets make a splendid servant, a bad
master, and a worse religion; then we have good news: displacing most,
probably all, of our oil not only makes sense, but also makes money, without even
counting any reductions in oil’s hidden costs.
This view reflects four insights from basic economics (but not the pundits’
glib and corrupted slogans):
• High energy prices are neither necessary nor sufficient for rapid energy
savings. They’re not necessary because modern, well-deployed energy-saving techniques yield phenomenal returns even at low prices—
as the leading firms mentioned above are demonstrating quarter after
quarter. Nor are high energy prices sufficient by themselves to overcome force of habit: DuPont’s European chemical plants were no more
efficient than its U.S. ones despite having long paid twice the energy
price, because all the plants were similarly designed.167
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Firms like DuPont,
IBM, and STMicroelectronics are
cutting their energy
use per unit of output
by 6% every year with
1–3-year paybacks.
That’s among the
highest and safest
returns anywhere.
High energy prices
are neither necessary
nor sufficient for
rapid energy savings.
162. Lovins & Lovins 1997.
163. DuPont’s goal “is to
never use more energy no
matter how fast we grow”;
it has cut greenhouse gas
emissions by 67% since
1990 without profits’ suffering (Wisby 2004). Raising
output 35% while cutting
energy use 9% has saved
~$2 billion (A.A. Morell,
personal communication,
11 June 2004).
164. Dr. Florentin Krause
says each $20 bill is actually more like 2,000 pennies.
As field practitioners, we
prefer another simile: saving energy is like eating an
Atlantic lobster. There are
big, obvious chunks of tender meat in the tail and the
front claws. There’s also a
roughly equal quantity of
tasty morsels that skilled
and persistent dissection
can free from crannies in
the body. They’re all tasty,
and they’re worth eating—
scraps, broth, and all.
165. For a critique of the
foundations of many contemporary economic constructs, see DeCanio 2003.
As a senior staff member of
President Reagan’s Council
of Economic Advisors, his
cogent arguments carry
special weight. See also
McCloskey 1996.
166. Lovins 2004.
167. See next page.
27
Oil Dependence
Four insights from
basic economics
(continued):
167. J.W. Stewart (DuPont),
personal communication, 9
October 1997. All these
examples are readily explicable in economic terms
and described in economic
literature dealing with organizational behavior, transaction costs, information
costs, etc. (Lovins & Lovins
1997, pp. 11–20). Our point is
rather that too many policymakers and analysts, whose
economic tools may not be
sharp enough, tend to treat
market failures as ad hoc
add-ons—contrived to
explain supposedly minor
departures from an underlying near-perfect market. In
reality, market failures can
be considered more important than market function,
not just because they’re so
pervasive, but because correcting them first is the main
opportunity for innovation
and profit.
168. Hawken, Lovins, &
Lovins 1999, p. 254.
169. For example, much of
the 1979–85 energy efficiency revolution was driven by energy price decontrols long and rightly urged
by economists, launched by
President Carter, and
accelerated by President
Reagan.
170. Namely, 2.9%/y, vs.
4.6%/y at the record-high
and rising energy prices of
1979–85.
171. As inventor Dean
Kamen puts it.
28
Could less oil dependence be not only worthy but also profitable? Beliefs that hold us back
• Ability to respond to price is more important than price itself.
During 1991–96, people in Seattle paid only half as much for a kilowatt-hour of electricity as did people in Chicago, but in Seattle they
saved electricity far faster (by 12-fold in peak load and 3,640-fold in
annual usage), because the electric utility helped them save in Seattle
but tried to stop them from saving in Chicago.168 That logical outcome
is the opposite of what the bare price signals would predict.
• Energy prices are important, should be correct,169 and are a vital way
to interpret past behavior and guide future policy; yet price is only
one way of getting people’s attention. U.S. primary energy use per
dollar of real GDP fell by nearly 3% a year170 in 1996–2001 despite
record-low and falling energy prices—evidently spurred by factors
other than price.
• Some theorists, their vision perhaps strained by years of peering
through econometric microscopes, suppose that the potential to save
oil in the future can be inferred only from the historic ratio of percentage changes in oil consumption to percentage changes in oil price.
This “price elasticity of demand” is real but small, though it gradually
rises over many years. In fact, it says little about what’s possible in the
future, when technologies, perceptions, and policies may be very different. Supposing that the nation can never save more energy than historic price elasticities permit, and that we must interpret and influence
choice only through price signals, is like insisting a car can be steered
only in a straight line chosen by staring in the rear-view mirror. Such
drivers, who pay dangerously little attention to what’s visible through
the windshield, risk getting slammed into by speeding technological
changes.
We therefore take economics seriously, not literally, and we blend it with
other disciplines to gain a more fully rounded picture of complex realities.
From this perspective, Americans still use so much oil mainly because
they’ve been inattentive to alternatives and overly comfortable in the
assumption that optimal progress will automagically171 be provided by a
competitive market. Americans have also been too ready to accept dismissive claims, often from those whom change might discomfit, that any
improvement will be decades away, crimp lifestyles and freedoms, and
require intrusive interventions and exorbitant taxation. Such gloom rejects
the rich American tradition of progress led by business innovation.
Consumer electronics every month get smaller, better, faster, cheaper;
why can’t anything else? And although cars normally last 14 years and
are getting even more durable, why can’t we accelerate the turnover of
the fleet—thereby opening up vast new markets for earlier replacement—
while revitalizing the factories and jobs that make the cars?
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Whatever exists is possible
Americans are starting to discover this different reality. Exhibit A is
Toyota’s 2004 Prius hybrid-electric midsize sedan (Fig. 4). Secretly developed and brought to market without government intervention, it’s rated
at 55 mpg, more than twice the fuel efficiency of the average 2003 car on
the U.S. market,173 yet it costs the same—between Camry and Corolla in
size and price, but more value-loaded:174 it has more features, and protects
you just as well.175 Sluggish? “Impressive performance, with unexpectedly
quick acceleration,” noted a tester for Motor Trend, which measured 0–60
mph in 9.8 seconds.176 Unpopular? Ever since it went on sale in October
2003, the Wall Street Journal has listed it as the fastest-selling automobile
in the United States, flying off dealers’ lots in 5–6 days—after 3–12
months on a waitlist. Car and Driver named Prius among its “Best Ten.”
It was Automobile’s Design of the
Year and the Society of Automotive
Figure 4: Toyota’s 2004 Prius
The Prius is a 5-passenger, near-zero-emission,
Engineers’ magazine Automotive
55-mpg, ~$20,000 midsize sedan
Engineering International’s Best
Engineered Vehicle. At Detroit’s
January 2004 North American
International Auto Show, the auto
journalists’ panel straightfacedly
named the Japanese-made Prius
North American Car of the Year.
Today’s doubledefficiency hybrid cars
illustrate how
dramatic oil savings
can be achieved
without sacrifice,
compromise, or even
appreciable cost.
172
172. When introduced in October 2003, the 2004 Prius ’s Manufacturer’s Suggested Retail Price (MSRP) before destination charges was $19,995, the same as
when Prius first entered the U.S. market in 2001. On 3 April 2004, Toyota raised Prius ’s MSRP to $20,295 as part of a ten-model general price increase to offset
the stronger yen. The 2005 model released 14 September 2004 has a base price of $20,875—triple the increase for other models. Most dealers still report
~6–12-month waitlists.
173. At 55 adjusted EPA mpg, the 2004 Prius is 107% better in ton-mpg than the best midsize model or 113% better than the average of all midsize and compact models. (Based on some contractors to the Big Three/federal PNGV [Partnership for a New Generation of Vehicles] program, Congress’s Office of Technology Assessment predicted in 1995 that hybrids could do this, and DOE established that as its program goal for hybrids: OTA 1995, pp. 175–176.) In size,
Prius is toward the lower end of the midsize range, but through innovative design, it has more cargo space than many midsize sedans. Like all hybrid cars,
Prius must be properly driven to approach or exceed its EPA-rated fuel economy; failure to do so typically yields in the low 40s of mpg (Rechtin 2003). “Pulse
driving” is recommended—accelerating rather rapidly, because a brief high engine load uses less fuel than a prolonged low engine load, and braking well
ahead of a stop to maximize brake regeneration. Prius can then reach or exceed 55 mpg (one driver got 70 mpg, albeit at <40 mph). Toyota USA’s executive
engineer reports that his 2004 Prius, driven with his expert knowledge, averages 53–55 mpg (Rechtin 2003), confirming that the issue is one not of automotive
technology but of driver re-education. (It’s too early for population-based data on the 2004 Prius, but the Prius-expert at (John’s Stuff 2004) shows that his
2001 Prius, rated at 48 mpg, has averaged 45.4 over 59,827 miles—close agreement in view of his Minnesota location’s predominance of cold weather, which
hurts regeneration, and snow tires, which cost 1–3 mpg.) Similar variations are observed among Honda Insight hybrid drivers, who can fall far short of the
rated 62–64 mpg if they ignore the tall gearing and upshift light. But without correcting for minor model differences or snow tires, the self-selected drivers at
www.insightcentral.net report lifetime averages of 63.0 mpg for 77 owners (the senior author of this study gets 63.4) and 61.92 mpg for 295 owners. Of course,
non-hybrid cars with average U.S. drivers also typically underperform their adjusted EPA mpg by ~5–15%; what’s different with hybrid cars is the technical
reasons—and the hybrids’ greater ability to approach or achieve their mpg ratings if well driven.
174. Specifications, features, and options at Toyota 2004b. The fullest option package adds $5,245 and offers many features absent from the costlier top-ofthe-line Camry option package. Hybrid cars in 2003 got a $2,000 federal tax deduction (which phases down starting in 2004), plus a tax credit in a half-dozen
states (www.fueleconomy.gov, undated). Japan currently offers a ~$2,300 rebate to hybrid-buyers (originally worth ~$3,500 at the exchange rate then prevailing), and subsidized 31,000 hybrids during 1998–2001.
175. The crash “star” ratings at NHTSA 2004, frontal-crash driver/passenger and side-impact front/rear, are 5/4/4/4 (5-star is the best possible rating). Prius
in 2003 was 4/4/3/3; the 2004 Corolla and Camry are respectively 5/5/4/4 and 4/4/4/3. All these 2004 ratings are considered excellent. The 2004 Prius also has a
good rollover rating (4-star, 13%, no tip), matching the best-rated MY2004 SUV on the market.
176. Some initial Toyota remarks claimed 10.1 s (4.9s for 30–50 mph) with a 103-mph top speed. An & Santini (2004) and current official Toyota presentations
say 10.5 s, and emphasize the 9% increase in interior volume, to 110 cubic feet, from the 2001 Prius.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
29
Oil Dependence
Whatever exists is possible
If all U.S. cars
(not light trucks)
were Priuses today,
they’d save
15% more oil than the
U.S. got in 2002 from
the Persian Gulf.
Its coveted Motor Trend “Car of the Year” award came with this editorial
comment:177
177. Italics added. All
Motor Trend quotations in
this paragraph are from
December 2003 issue and
Motor Trend 2003a.
The Prius is a capable, comfortable, fun-to-drive car that just happens to get
spectacular fuel economy. It also provides a promising look at a future where
extreme fuel-efficiency, ultra-low emissions, and exceptional performance will
happily coexist. That makes it meaningful to a wide range of car buyers.
Toyota “says it’ll make money on each one it builds,”178 and “a large portion of its model lineup will offer hybrid power by 2007.” President Fujio
Cho plans to make 300,000 hybrids a year (about half for the North American market179), boosted U.S. production by 31% and has been asked for
more,180 and licensed the powertrain technology to Nissan and Ford181 and
offered it to GM and DaimlerChrysler.182 Economic theory notwithstanding, you can pick up several $100 bills a year off the ground at your local
gas station without extra cost. If all U.S. cars (not light trucks) were Priuses
today, they’d save 15% more oil than the U.S. got in 2002 from the Gulf.
178. Some commentators,
including Motor Trend, initially speculated that the
Toyota, whose ¥1-trillion 2003 profits and ¥2-trillion liquidity lets it place
2004 Prius was a lossleader, as it probably was
multiple technology bets, is widely considered at least three or four years
in its early years; Detroit
ahead of Detroit with this third-generation hybrid technology. As the
experts estimate added
production cost at about
world’s number-two automaker (it passed Ford in 2003), with market
$2,500–4,000. But Toyota
capitalization exceeding that of the Big Three combined, and as arguably
denies a loss; Executive
Vice President Yoshio
the world’s premier manufacturing company, Toyota can price its cuttingIshizaka confirmed 5
edge innovations aggressively and deploy them in many sizes, shapes,
January 2004 that after
reducing hybrid-system
and segments:183
costs by more than 30%
since 1997, “Every Prius we
“This Prius is an industry sputnik,” says Jim SanFillippo, [executive VP] of
sell, we make a profit”
Automotive Marketing Consultants Inc., Warren, Mich. “The question is will
(Hakim 2004). Welch (2004)
there be a serious response, particularly from American competitors—because
says Prius didn’t turn profitable until MY2003, but
there should be. It is a serious, serious piece of technology.”
Chairman Hiroshi Okuda
was already saying Prius
was profitable in a 29 September 2002 Financial Times story. Reuters explains how: the 2004 Prius is assembled at a rate of one per minute on the same
Tsutsumi line as four other mass-production sedans, where every other car made is now a Prius (Kim 2003). The 2004 global Prius sales target of 76,000 was
recently increased as U.S. targets were boosted by 31%; first-month global sales of the 2004 model matched 2003’s total (Toyota 2003). An authoritative
source confirms that in early 2004, the 2004 Prius was incurring an extra production cost corresponding to a ~$4,000-higher (2003–04 $) retail price, but that
this marginal MSRP is intended and expected to fall to $2,000 (~2007 $): D. Greene, personal communication, 25 March 2004. Normal industry cost structure
would imply marginal manufacturing costs roughly half as big, but a Toyota official has estimated they’re ~$2,500–3,000, yielding a reduced but still positive
factory margin, rumored to be ~$1,100 (~2003 $): Peter 2003. These figures depend on accounting conventions: Toyota probably assumes that every Prius is a
sale it wouldn’t otherwise make, and undoubtedly amortizes the powertrain R&D over many future models. It’s not clear whether the attractive margins on
options are included. And these Prius data are consistent—if adjusted for powertrain complexity and vehicle size—with current Honda Civic Hybrid estimates of at least $3,000 extra MSRP, intended to be reduced to ~$1,500.
179. Nihon Keizai Shimbun reported (AFP 2004) in March 2004 that Toyota plans a North American hybrid version of its best-selling Camry in 2006:
M.P. Walsh 2004 (pp. 32–33; pp. 45–48) reports this as a Toyota announcement.
180. Mid-May 2004 press reports indicate a 150% rise in Prius sales in April 2004 vs. April 2003, and say Toyota expects to sell 50,000 Prius cars in 2004 vs. its
original target of 34,000. Lyke (2004) reports a monthly AutoVIBES report from Harris Interactive and Kelley Blue Book finding that 17% of U.S. new-car buyers have already changed their minds about which vehicle to buy because of fuel prices, 21% are strongly considering models they hadn’t previously considered, and 15% say they’d strongly consider more fuel-efficient vehicles if gasoline prices rose just 25¢/gal (Automotive.com 2004).
181. The license, announced (probably belatedly) in March 2004, is for 20 Toyota patents on hybrid systems and control technology, but does not include the
use of hybrid powertrain components as the 2002 Nissan licensing agreement does (Toyota 2004), and may not be as advanced as the 2004 Prius.
182. Peter 2003.
183. Italics added. Freeman 2003.
30
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Oil Dependence
Whatever exists is possible
Figure 5: Four hybrid SUVs
From L–R, they are: 2005 Ford Escape, 5 seats, 32 mpg in 2WD, 1,000-lb towing, ~$27,000; 2005 Lexus RX 400h, 5 seats, 3.3-L V-6,
270 effective hp, 0–60 mph in <8 s, range >600 miles, SULEV, AWD option, 1,000-lb towing, ≥28 mpg, ~$39,000;
2005 Toyota Highlander, 7 seats, similar characteristics, 3,500-lb towing option; and the 2005 Mercedes Vision Grand Sports Tourer,
6 seats, 314 hp, 0–62 mph/6.6 s, 155 mph, 33 mpg.
Detroit is of course striving to emulate this brilliant achievement of
Japanese engineering. GM, a leader in hybrid buses, postponed its general-market hybrid cars to 2007, dismissing them as an “interesting curiosity” that does not “make sense at $1.50 a gallon,” 184 and is introducing only
slightly more efficient “mild-hybrid” pickups in 2004 while focusing on
longer-term fuel-cell development.185 But Ford is launching in October
2004 the 2005 Escape SUV (Fig. 5a) with a hybrid powertrain like Prius’s.186
It’ll offer roughly three-fifths better fuel economy than the 2004 201-hp
V-6 Escape’s 20 mpg—32 mpg with front-wheel or somewhat less with
four-wheel drive—yet provide comparable acceleration with only a 2.3-L,
4-cylinder Atkinson engine. But initial production is targeted for just
20,000 a year, one-sixth of the 2004 Prius’s volume, and priced at a few
thousand dollars’ premium.187 A fancier Mercury Mariner model will follow about a year later, then a midsize car after 2006.188 Analysts differ on
whether Ford can make money on these early hybrids. Clearly Detroit has
a lot of catch-up to do with its four planned SUV hybrids in 2005–07—
and its competitors are rapidly moving targets.
Four months after Ford’s hybrid Escape is launched, Toyota will unleash
brawnier versions of Prius’s powertrain: in February 2005, both the Lexus
RX 400h luxury hybrid SUV based on the RX 300,189 combining V-8-like
acceleration from a V-6 with the “fuel economy of a [4-cylinder] compact
car,” 190 and the Toyota Highlander 7-seat hybrid SUV (Fig. 5b).191 Honda is
meanwhile launching its peppy Accord V-6 hybrid midsize sedan in
autumn 2004.192 Hybrids’ powerful torque also boosts acceleration in
Honda’s beefy 2003 SU-HV1 concept SUV and CS&S roadster, GM’s 2003
Chevy S-10 concept pickup (which can “smoke” a Corvette193), and
184. Robert Lutz (GM Vice
Chairman/Product
Development), remarks at
the Detroit auto show, 6
January 2004 (Isidore 2004).
185. Hakim 2003. Originally
GM had planned to offer a
hybrid option on its MY2005
Saturn Vue small SUV. The
new plan would put it on
two MY2008 full-sized
SUVs, using a scalable system that fits in a 6-speed
transmission envelope and
is expected to boost mpg
by 25–35%. Meanwhile, late
in 2003 GM began offering
mild-hybrid Chevrolet
Silverado and GMC Sierra
pickup trucks to fleet customers; these hybrid-assist
models should reach other
customers in the third quarter of 2004, as will a
Chrysler Dodge Ram mildhybrid pickup. Such heavy
vehicles’ fuel savings in
gallons will be substantial.
186. Information from
Ford Vehicles, undated.
187. Information from Ford Vehicles 2004(a); a Prius-like $3,500 premium was estimated by WardsAuto.com (2004).
By mid-May 2004, 34,000 potential buyers had signed up for the Escape E-newsletter (Schneider 2004).
188. Crain 2004.
189. Information from Lexus, undated (downloaded 6 January 2004).
190. All data from Lexus, undated (downloaded 1 January 2004).
191. All data from Toyota, undated (downloaded 6 January 2004).
192. Its acceleration is expected to beat that of the 240-hp nonhybrid Accord.
193. GM 2003.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
By the end of 2005,
at least a dozen diverse models of hybrid-electric
cars, pickups, and SUVs
will be in American showrooms.
31
Oil Dependence
Whatever exists is possible
A self-interested
melding of advanced
technology with bold
business strategies
can present
extraordinary
opportunities for
private enterprise
to lead in making a
better world.
Mercedes-Benz’s six-seat diesel-hybrid Grand Sports Tourer (Fig. 5d), to be
launched in Europe early in 2005.194 Automakers will introduce hybrids in
middle and high-end products with greater amenity and performance
requirements and richer margins. But they won’t stop there.
As hybrids become cheaper, they’ll diffuse downmarket and become
ubiquitous. At the October 2003 Tokyo Motor Show, 10 of 25 cars featured
in Car and Driver were hybrids, by Daihatsu, Honda, Jeep, Lexus, Mazda,
Mercedes-Benz, Subaru, Suzuki, and Toyota, plus Nissan.
The 2004 Prius shows more than the competitive challenge to U.S. automakers. In the global marketplace, a self-interested melding of advanced
technology with bold business strategies can present extraordinary
opportunities for private enterprise to lead in making a better world.
Toyota didn’t wait for government to tell it what to do or pay it to innovate; it simply led.
Likewise in America, the 1977–85 “practice run” (p. 7) shows there’s
nothing mysterious about saving oil quickly and massively. It takes attention; leadership at all levels and in many sectors of society; a comprehensive and systematic but diversified and flexible approach; a spirit of
adventure and experimentation; enlisting cities and states as policy laboratories to speed up learning; openness to other countries’ technical and
policy innovations; and intelligent exploitation of today’s tripolar world,
where the private sector and civil society increasingly collaborate to fill
gaps left by government. In 1977–85 federal policy and industrial innovation succeeded beyond all expectations.195 But today, not all action need
await decisions by the national government. Our more dynamic society,
empowered by the Internet and its self-organizing networks, has far more
ways to get things done. Gridlock in Washington—captured by the
bumper-sticker “Progress: Opposite of Congress”—is no longer a reason
for inaction, but a spur to private-sector, civil-society, and state- and localgovernment experimentation and initiative. And we daresay America
could mobilize to get off oil even faster by drawing on the leadership
qualities and organizing skills of certain retired military personnel,
who understand why displacing oil defends the values we hold dear,
and how to execute that new mission decisively.
194. From MercedesBenz, undated (downloaded 7 January 2004);
Mercedes-Benz 2004;
DaimlerChrysler AG
2004; Germancarfans.com 2004.
195. Greene 1997.
The 1979 oil price shock
(Fig. 3, p. 15) certainly
helped too.
32
The rest of this report outlines how—by technology, business strategy,
and policy innovations—the United States of America can rise together to
the greatest challenge of this generation: winning the oil endgame in a
way that preserves and enhances our freedoms, our prosperity, and our
quest for common goals of substance and spirit, applying the problemsolving prowess that for two centuries has helped to make the nation and
the world better and safer.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
This Report
Structure and methodology
Pages 29–126 summarize an independent, transdisciplinary analysis of
four ways to displace oil:
1. Using oil more efficiently, through smarter technologies that wring
more (and often better) services from less oil (pp. 29–102).
2. Substituting for petroleum fuels other liquids made from biomass or
wastes (pp. 103–111).
3. Substituting saved natural gas for oil in uses where they’re interchangeable, such as furnaces and boilers (pp. 111–122). Note that gas
and oil, though sometimes found and thought of together, are utterly
different in geology, economics, industry, and culture.
4. Replacing oil with hydrogen made from non-oil resources (postponed
to pp. 228–242).
These options will be described first separately and then in integrative
combination, because together they can do more than the sum of their
parts. Efficiency options are presented for each end-use of oil. Each class
of oil-displacing technology is presented as two different portfolios:196
• Conventional Wisdom: Expectations broadly accepted by industry and
government; technologies on or soon to enter the market using thoroughly proven methods; surprisingly often already overtaken by the
best technologies already on the market.
• State of the Art: Best technologies sufficiently developed by mid-2004,
applying established principles and techniques, to be expected confidently as timely and competitive market entrants; well supported by
empirical data and industry-standard simulations; require no technological breakthroughs; not all off-the-shelf, but no longer heretical.
This study
transparently and
rigorously applies
orthodox market
economics and
measured performance
data to assessing
how much oil can be
displaced at what cost.
We present
two portfolios of
technologies and
contrast two implementation methods to
highlight key choices
and consequences.
The current public-policy approach to implementation might be
described as:
• Gridlock as Usual: Political will, chiefly focused on national policy and
driven by traditional constituency politics, can make modest, incremental advances comparable to those of the past two decades, but
cannot execute gamechanging shifts in the status quo.
In contrast, pp. 127–168 present the business case, and pp. 169–226 the
policy content, of:
• Coherent Engagement: Advanced technologies are rapidly and widely
adopted via innovative business strategies and a diverse portfolio of
innovative policy instruments with trans-ideological appeal, weaving
a rich tapestry of experiments by diverse actors.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
33
This Report
Structure and methodology
Readers might expect us to outline at least three197 internally consistent
scenarios for America’s path beyond oil. But the first is uninteresting:
• Drift: Conventional Wisdom technologies
plus Gridlock as Usual implementation;
We briefly mention it on pp. 180–181 as a recalibration of the base-case
forecast described below, just to clarify that some oil savings would occur
even without our policy portfolio. We pay slightly more attention to what
good policies can do even with incremental technologies:
• Let’s Get Started: Conventional Wisdom technologies
plus Coherent Engagement implementation;
But we focus mainly on the “best of both worlds”:
• Mobilization: State of the Art technologies
plus Coherent Engagement implementation.
None of these possible futures is a forecast, but we believe all are possible. Their divergent results show the importance of wisely choosing the
future we want, then fearlessly creating it.
Our calculations can be
reproduced on a hand
calculator or with a
simple spreadsheet. All
calculations are shown
and documented in the
Technical Annex.
196. We use this term in
this report to mean a cluster of internally consistent
assumptions that help to
understand the nature and
implications of choices.
This differs from the meaning in scenario planning:
Schwartz 1996.
197. We don’t explore the
off-diagonal matrix element
combining State of the Art
technologies with Gridlock
as Usual implementation
because those technologies are unlikely to be commercialized within that policy framework.
To ground our possible futures in policymakers’ day-to-day reality, each
technological option gauges potential oil displacement against the U.S.
economic activity and oil consumption projected for 2025 in the Reference
Case of the U.S. Energy Information Administration’s Annual Energy
Outlook 2004.198 EIA uses the National Energy Modeling System (NEMS),
which lacks the predictive power, modern technological assumptions, and
structural flexibility needed for sound business planning, but is widely
used by government. Our analysis uses NEMS’s outputs (e.g., forecast oil
use by each sector and each class of end-use device in 2025) and inputs
(e.g., how briskly new light vehicles sold in 2025 will accelerate, so we
can match our assumed vehicle designs to that performance). However,
we have chosen not to use the NEMS model itself, nor any other large
computer model of the energy system. (Econometric models, though
widely used, would be especially misleading because they rely on historic
coefficients that our proposed technological and policy changes are meant
to transform.) Rather, our calculations have been kept so simple and
scrutable that they can all be reproduced on a hand calculator or with a
simple spreadsheet. All calculations are shown and documented in the
Technical Annex, so they can be checked and so readers who prefer other
assumptions can plug in their own. We believe that for modeling the
long-term energy system, less is more, that it’s better to be approximately
right than precisely wrong, and that simplicity and transparency trump
complexity and opacity.
198. EIA 2004.
34
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Structure and methodology
This Report
As a baseline for energy comparisons, we mainly use the year 2000, both
because its statistics are available and stable (those for 2002 and even 2001
are still subject to revision at this writing in mid-2004 199) and because 2000
was a broadly typical year for most energy statistics.
Though we remain mindful of market failures and the importance of correcting them, our economic analysis rests on orthodox market principles
and economic methods—with one exception. To avoid using a large and
opaque model, we haven’t performed a general-equilibrium simulation to
test how far the strong efficiency improvements we describe would
undermine their own viability by reducing the prices of energy or of energy services.200 However, our conclusions are made more robust by multiple countervailing forces, including: About half of the State of the Art enduse efficiency potential using 2004 technologies can compete even at the
lowest oil prices (Fig. 3) that might be expected to result even from its full
global adoption (its use just in the United States, a fourth of the world
market, would cut oil prices by only one-fourth as much); wide global
adoption of strong energy efficiency could take about as long as depletion
of low- and intermediate-cost oil outside the Gulf, further focusing
OPEC’s market power; China, India, and others will meanwhile want
more oil; once adopted, the efficient vehicles, boilers, insulation, and other
oil-using devices are unlikely to be switched back to inefficient ones; and
if oil gets too cheap, it could always be taxed. Of course, if we’re wrong,
a durable oil surplus would be a nice problem to have.
We test cost-competitiveness against EIA’s January 2004 projections of real
energy costs in 2025, which are 14% above 2002’s real costs for world oil,
8% for gasoline, and 19% for retail natural gas. However, EIA’s 2025
U.S. Refiner’s Acquisition Cost, $26.08/bbl in 2000 $, is well below the
~$35–50/bbl (2004 $) world market prices prevailing in summer 2004.
EIA’s energy cost projections aren’t market price forecasts or expected
values; they assume that weather, climate, inventories, regulations,
geopolitics, and other background conditions remain “normal.” We use
them because EIA considers them consistent with its demand projections—our base case.
Our Conventional Wisdom technology options rely heavily on prior studies
by authoritative industry, government, and academic teams. Many of
these studies are sound, available, documented, and adequate. Where
such prior work is unavailable or insufficient, our own State of the Art
analysis is documented in the Technical Annex. Our limited analytic
resources are focused mainly on the seven biggest terms, each at least 6%
of 2025 U.S. oil use (Fig. 6), treating only briefly the small terms that make
up the last 6%.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
For modeling the
long-term energy
system, less is more,
it’s better to be
approximately right
than precisely wrong,
and simplicity
and transparency
trump complexity
and opacity.
199. For example, EIA’s
Annual Energy Review
showed 2000 petroleum
consumption 1.2% higher in
its 2002 edition (EIA 2003c)
than in its 2000 edition (EIA
2001c). We prefer to wait a
few years until the numbers
stabilize.
200. Such a calculation
would entail many assumptions and great complexity—akin to the inverse of
the challenge of calculating
macroeconomic effects of
increases in oil price,
where results vary substantially depending on model
structure and assumptions
(IEA 2004b, pp. 5–6), especially about exchange rates
and monetary policy.
Unmodeled effects, such as
changes in business and
consumer confidence and
in gas pricing, can be
important (IEA 2004b).
35
This Report
Structure and methodology
Figure 6: The end-uses of U.S. oil in 2000 (model input, 1% above actual) and projected for 2025, by energy content, according to the U.S. Energy Information Administration (2004, Reference Case), scaled by the
area of each pie. We compare all ways to save or substitute for oil with EIA’s 2025 projection.
2000 U.S. oil end use (19.7 Mbbl/d)
tria
l fe
ed
light trucks
16.9%
sto
11. cks
7%
el 12.7%
industrial fu
ild
in
7.7
heavy trucks
11.3%
transportation
lubricants
0.5%
%
s
bu
gs
cars
23.3%
ne
p la
a ir %
8 .1
buses & transit 1.2%
rail freight 1.3%
ships & boats
3.7%
military
1.5%
The main 2000–25 changes are general growth and a further shift from cars to light trucks.
The 2.9→1.5% of oil used to make electricity (see p. 98) is allocated here to the sectors that use the
electricity. The graphed end-uses of petroleum products supplied (including LPG) conventionally include
ethanol and MTBE 201 oxygenates equivalent in 2000 to 0.25 Mbbl/d crude-oil-energy-equivalent or 1.3% of
oil consumption, plus biodiesel equivalent to 0.02% of diesel fuel.202 EIA projects oxygenates to become
0.28 Mbbl/d or 1.0% of oil consumption by 2025.203 Light trucks include here both passenger (Class 1–2) and
commercial (Class 3–6, using 0.29 Mbbl/d in 2000 and 0.42 Mbbl/d in 2025); heavy trucks are Class 7–8.
Of military petroleum use, 94% in FY2000 was for transportation, mainly aviation, and nearly all the rest for
buildings. Construction and agricultural equipment is classified as industrial use, not transportation.
Lubricants, divided equally between industrial use and transportation in 2000 and assumed
likewise in 2025, use 1% of oil. Within feedstock end-use, 3.3% of total oil use in 2000 and 3.1% in 2025 is
asphalt and road oil. Totals may not add exactly due to rounding.
ind
us
tria
l fe
2025 U.S. oil end use (28.3 Mbbl/d)
ed
sto
10. cks
4%
el 11.2%
industrial fu
%
5.7
cars
17.2%
heavy trucks
12.3%
transportation
lubricants
0.5%
nes
airpla
7.8%
s
ing
ild
bu
light trucks
28.5%
transportation
72.8%
A doubled economy
and an assumed
huge shift to inefficient light trucks,
such as SUVs,
drive the Energy
Information Administration’s forecast of
44% more U.S. oil use
in 2025 than in 2000
(shown by making
the 2025 pies that
much bigger in area).
Light and heavy
trucks cause
70% of the increase.
We focus mainly
on trucks, cars,
airplanes, industrial
fuel and feedstocks,
and buildings
because they use
94% of the oil
in 2025.
us
transportation
67.8%
ind
ships
& boats
2.8%
rail freight 1.0%
buses & transit 1.1%
military
1.5%
Source: RMI analysis from EIA 2003c; EIA 2004; January 2004 NEMS database, kindly provided by EIA. NEMS doesn’t explicitly account for such small terms as motorcycles, snowmobiles, and all-terrain vehicles, but the bottom-up composition shown here fits the transportation and all-sector totals within less than 1% cumulative error.
201. Methyl tertiary butyl ether.
202. In addition, vehicular fuel equivalent in 2000 to 0.3% of gasoline was supplied by liquefied petroleum gas (68%),
compressed natural gas (27%), liquefied natural gas (0.02%), methanol (0.003%), and 85% or 95% ethanol (0.02%): EIA 2002.
203. This is because ethanol is due to replace MTBE in 17 states “over the next few years” (EIA 2004, p. 256),
and ethanol is twice as effective an oxygenate per unit volume (EPA 1998).
36
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
This Report
Structure and methodology
We have also focused chiefly on
the terms accounting for most of
the demand growth (Fig. 7).
Figure 7: Transportation Petroleum Use by Mode (1970–2025)
Of the growth EIA projected in January 2004 for U.S. oil consumption during
2000–25, graphed here by the U.S. Department of Energy, 55% was for light
trucks, 15% heavy vehicles, 7% aircraft, and only 3% automobiles.204
Total oil used for transportation surpassed domestic oil production
(crude oil plus lease condensate) starting in 1987, and is projected to be
twice domestic production by 2010.
Conservatisms and conventions
We believe our findings are conservative—tending to understate the
quantity and overstate the cost of oil-displacing alternatives—for four
main reasons:
1. We assume little future technological innovation—just what’s already
in the commercialization pipeline. This is likely to understate future
opportunities dramatically, much as if we sought to solve problems in
2004 by using only the technologies already being commercialized in
1979 (like IBM’s first personal computer, which came on the market
two years later). Though our broad conclusions don’t depend on the
biggest breakthroughs now emerging in oil-displacing technologies
(such as cheap ultralight autobodies and cheap durable fuel cells),
we do consider those at least as plausible as any conceivable further
advances in finding and lifting oil, and probably far more so.
2. We uncritically adopt EIA’s official projection that in 2025, 20% more
Americans—348 million people—will use 40% more energy and 44%
more oil than 289 million Americans used in 2002. This extrapolation
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
2025
2020
2015
2010
2005
2000
1995
1990
1985
1980
1975
1970
million barrels per day
Our emphasis on empiricism leads
us, wherever possible, to rely upon
and document actual measure22
ments rather than relying on theo20
actual
projected
air
retical projections. (Similarly,
18
16
where results seemingly contrary
domestic
heavy
14
to economic theory—such as
vehicles
production
marine
12
rail
expanding rather than diminishing
10
light trucks
returns to investments in energy
8
productivity—have been solidly
6
off-road
established by empirical practice,
4
cars
we don’t reject them in favor of
2
theory.) And we have followed
0
Aristotle’s counsel 205 that “it is a
mark of educated people, and a
year
Source: Transportation Energy Data Book: Edition 23, DOE/ORNL-6970, October 2003; and EIA Annual Energy Outlook
proof of their culture, that in solv2004, January 2004.
ing any problem, they seek only
Our findings are
so much precision as its nature
conservative. We’ve
permits or its solution requires.” Energy analysis, especially looking a
measured potential
quarter-century ahead, is an uncertain art, so we strive to avoid implying
oil displacements
spurious precision.
against the government’s generous
projection of demand
in 2025. We haven’t
included technologies
still to be developed
nor important nontechnological ways to
save oil, such as
not subsidizing nor
mandating sprawl.
204. From EERE 2003. The
car/light-truck data, which
EIA’s Annual Energy
Outlook doesn’t show, were
confirmed from the NEMS
database. Light trucks
include both passenger
(Class 1–2) and commercial
(Class 3–6).
205. A paraphrase combined from Nicomachean
Ethics I:3.24 (Berlin 1094b)
and I:7.26 (Berlin 1098a).
37
This Report
Conservatism #2
(continued):
206. TTI 2003.
207. Literature summarized
by Lovaas (2004) convincingly shows that continuing
to expand land-use twice
as fast as population is not
necessary, desirable, or
economic, and that more
thoughtful land-use patterns can reduce cost and
crime, increase social
cohesion and economic
vitality, reduce travel needs
by tens of percent (even
more with the drop in
“induced travel” when
fewer roads must be built),
and increase real-estate
values and profits.
208. Shifting from cars to,
say, transit buses may or
may not save fuel, depending on efficiencies and load
factors: e.g., ORNL 2003,
Table 2.11.
209. A Dutch pilot study
suggests point-to-point
flight patterns could serve
the same origin-destination
network with ~17% less
fuel: Peeters et al. 1999,
summarized in Peeters,
Rietveld, & Schipper 2001.
210. Hawken, Lovins, &
Lovins 1999, Ch. 14.
211. See
www.cybertran.com.
212. Such as the software
and hardware suite
described at TransDecisions 2003. A further
step could be applying to
civilian logistics the intelligent-software-agents
approach developed in
the Defense Advanced Research Projects Agency’s
Advanced Logistics Project:
Carrico 2000.
213. These are being commercialized by German
(including Lufthansa) and
Russian consortia: examples are at Aerospace
Technology, undated and
RosAeroSystems, undated.
Conservatisms and conventions
reflects a far from deprived future: a 96% higher GDP, 97% bigger personal disposable income, 22% larger labor force, 80% higher industrial
output, 81% more freight trucking, 41% more commercial floorspace,
and 25% more and 6% bigger houses. Each person will drive 33% and
fly 48% more miles. Total light-vehicle miles rise 67%. Average
horsepower rises 24% for new cars and 18% for new light trucks,
while light-vehicle efficiency improves 3 mpg for new models and 1.2
mpg, or 6%, for the operating fleet, and light-vehicle annual sales
grow 27%. EIA’s forecast also assumes a 29.4% (1.5%/y) drop in primary energy intensity during 2002–25, encouraged by slightly costlier
energy (except electricity, which gets 4% cheaper). Every indicator of
consumer choice and material standard of living rises steeply, though
one might wonder about such quality-of-life indicators as inequities,
social tensions, dissatisfaction, and traffic congestion (which in 2001
wasted 3.6 billion hours and 5.7 billion gallons, worth $70 billion, in
75 U.S. urban areas206). We also assume that the required expansions of
infrastructure (driving and flying space, refineries, pipelines, etc.) will
all occur smoothly, undisturbed by any unbudgeted costs or ill effects.
3. We omit many oil-saving options that are individually often large and
collectively immense, and we don’t analyze some others that are individually small but collectively significant. The main transportation
options we deliberately omit, many important, include:
• Improving land-use (e.g., by not subsidizing nor mandating
sprawl, or by smart-growth initiatives) so people needn’t travel so
much to get where they want to be, or ideally are already there;207
• Telecommuting and other substitutions of telecommunications for
physical mobility;
• Shifting passenger transportation modes, e.g., from road or air to
rail, car to public transit,208 or driving to walking and biking;
• Increasing passenger load factors, except by HOV lanes and normal
airline methods;
• Restructuring the airline industry (through authentic gate and slot
competition) to permit fair competition between hub-and-spokes
airlines and a Southwest-Airlines-like hubless point-to-point pattern that reduces unnecessary aircraft movements, travel time, misconnections, hub congestion, and oligopoly rents; 209
• Innovative public transit vehicles such as Curitiba-style 210 express
buses or Cybertran®-style211 ultralight trains;
• Improved freight logistics,212 and innovative vehicles such as
containerized airships,213 although we do include continued gradual expansion of road/rail freight intermodality.
214–9. See Box 5 on p. 39.
38
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
This Report
Conservatisms and conventions
4. Instead, we adopt only options that provide EIA’s projected 2025
mobility “transparently” to the user, with no change of lifestyle or loss
of convenience—other than those entailed by the congestion implicit
in EIA’s enthusiastic traffic forecasts. Similarly, beyond any changes
implicit in EIA’s growth forecast, we assume no shifts in where, when,
or how people use buildings (just more of everything including
sprawl); no savings in the materials that industry processes and fabricates (due to materials-frugal and longer-lived product design or
closed materials loops, except plastics recycling); and no changes in
when people choose to use electricity (even though price signals to do
so, and “smart meters” and other technologies to make demand
response convenient, are rapidly emerging).
We adopt only
options that provide
EIA’s projected
2025 mobility
“transparently”
to the user, with no
change of lifestyle or
loss of convenience.
Using the technical conventions in Box 5, we next present our analysis
and findings of what each of the four oil-displacing options can do by
itself in each of the two technology suites. We’ll combine them successively, showing how each leaves less oil to be displaced by later options (pp.
123–125, 227–230). Having contrasted Conventional Wisdom and State of the
Art technologies, each incorporating all four oil-displacing options in
integrated form, we’ll finally discuss their implementation, global context,
and strategic implications.
5: Conventions
(details are described in the Technical Annex, Ch. 4)
Units and conversions: We use customary U.S.
units, plus international (metric) units in parentheses where readers from countries other than
the United States, Liberia, and Myanmar are
most likely to want them. Year is abbreviated “y”,
day “d”, hour “h”, second “s”. Tons are U.S.
short tons (2,000 lb); tonnes are metric (1,000 kg).
Thousand (103) is “k”, million (106) is “M” (the
worldwide scientific, not the U.S. engineering,
convention), and billion (109) is “b”. We use
industry-standard conversion constants, including EIA’s214 for fuels. By industry and EIA convention, we express fossil-fuel prices, energy content, and efficiency at the Higher Heating Value
(which includes the energy needed to evaporate
water created by combustion), but we use the
Lower Heating Value for hydrogen (120 MJ/kg).
Economic assumptions: Unless otherwise noted,
all monetary quantities are converted to constant
2000 U.S. dollars (2000 $) using the GDP implicit
price deflator.215 All discounting—to present value
or to levelize cost streams, where noted—applies
a 5%/year real discount rate. Pages 189–190 discuss the spread between this conservatively high
social rate and much higher implicit consumer
discount rates; pp. 16 and 101, risk-adjustment.
(The federal government uses 3.29%/y for most
long term projects, and 3.0%/y for its own energy-efficiency investments.) Prices in our analysis
exclude sales and ownership taxes on vehicles
and all taxes on retail fuels, because those taxes
are transfer payments, not real resource costs.
(continued on next page)
214. EIA 2003c, Appendix A.
215. BEA 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
39
This Report
Conservatisms and conventions
5: Conventions (continued)
Data sources: U.S. energy data not otherwise
cited are from the U.S. Energy Information
Administration (EIA, www.eia.doe.gov), typically the Annual Energy Review 2002, Annual
Energy Outlook 2004, or sectoral yearbooks.
Transportation data not otherwise cited are from
Oak Ridge National Laboratory’s Transportation
Energy Data Book 23 (2003, ORNL-6970,
www-cta.ornl.gov/data), which relies heavily on
U.S. Department of Transportation data.
Road-vehicle efficiency metric: Unless otherwise stated, we measure fuel economy in the
“adjusted” USEPA (combined city/highway driving
cycle) miles per U.S. gallon (mpg) of gasoline or
equivalent—the same figure used for sales stickers, but 10%/22% below the city/highway “laboratory” mpg measured in EPA testing and used
for Corporate Average Fuel Economy (CAFE)
regulation. (When calculating the corresponding
oil usage or savings, we use EIA’s “degradation
factor” to obtain “on-road” mpg.) Fuel intensity
is expressed in gal/mi or L/100 km (liters per 100
kilometers).
Hydrocarbon definitions: As in EIA statistics,
“petroleum” is the sum of crude oil, lease condensate, and natural gas plant liquids. Thus
liquefied petroleum gas (LPG) is “petroleum”
whether it was derived from producing oil or natural gas, and whether it is used as a fuel or as
a feedstock. “Oil” is used in this report to refer
to “petroleum” or “refined petroleum products”
or both, according to context (production or use).
We include all non-fuel (feedstock) oil consump216. EIA’s 2000 “oil“ consumption includes 0.25 Mbbl/d (in crude-oil
energy equivalent) of biomass-derived ethanol and natural-gasderived MTBE. We obtain the sources of MTBE in 1996, which we
assume also for 2000, from Wang 1999, who says that over 90% of
MTBE’s isobutylene and methanol components originate in natural-gas
processing plants rather than from petroleum. EIA thus understates
the degree of biomass- and natural-gas-for-oil substitution that has
already occurred to produce these oxygenates.
40
tion. We adopt EIA’s convention of accounting for
oxygenate production and consumption as petroleum-based (even though most of it is not 216),
with one noted exception on p. 1 (see note 15).
Petroleum processing: U.S. statistics measure
oil by volume (1 barrel ≡ 1 bbl ≡ 42 U.S. gallons =
159 L), but refining heavy crude oil into lighter
refined products expands its volume while consuming energy and money. We convert from
end-use fuel savings back to equivalent crudeoil refinery inputs not by relative volume but by
using their EIA-compiled relative energy content
as the best measure of utility. We don’t adjust
that conversion for refineries’ energy use
because that’s counted as part of industrial
energy consumption. We conservatively convert
the cost of saved fuel back to 2000 Refiner’s
Acquisition Cost (RAC) solely on the short-run
margin, i.e., by deducting only the 2000-average
cash operating costs needed to refine the crude
oil and deliver it to the point in the value chain
where the savings occur. (These cash operating
costs include EIA’s estimate of incremental
investment to comply with low-sulfur regulations, but aren’t adjusted for any deferral or
avoidance by reducing demand.) This valuechain addback thus calculates the RAC of crude
oil that would have produced and delivered that
saved fuel in the short run from existing industry
physical assets, taking no credit for avoided
future refining or delivery capacity. Thus an RAC
of $26.08/bbl (2000 $), EIA’s projected 2025 crudeoil price, breaks even against saving 76¢/gal pretax retail gasoline. The difference between 76¢
and the pretax retail price reflects all embedded
costs from refinery to filling station. Our oil displacements could avoid further investment for
new refineries, pipelines, delivery fleets, etc.
For this reason, and because our conversion
isn’t risk-adjusted, it understates oil displace(continued on next page)
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
This Report
Box 5: Conventions (continued)
ments’ implied value. Our supply-curve graphs
show both RAC $/bbl and retail ¢/gal (of the fuel
being displaced—gasoline, diesel fuel, or jet
fuel), but due to the value-chain addback and
the energy-per-unit-volume conversion, their
relationship isn’t simply the standard conversion
factor of 42 gal/bbl.
Cost of Saved Energy: We express the cost of
saving a unit of oil as a Cost of Saved Energy
(CSE), calculated at the point in the value chain
where it occurs, then converted (as just
described) to equivalent short-run RAC per barrel of crude oil. We calculate CSE using the
standard formula developed at Lawrence
Berkeley National Laboratory. This divides the
marginal cost of buying, installing, and maintaining the more efficient device by its discounted
stream of lifetime energy savings. Using the
standard annuity formula, the dollar cost of saving 1 bbl then equals Ci/S[1–(1+i) –n], where C is
installed capital cost ($), i is annual real discount rate (assumed here to be 0.05), S is the
rate at which the device saves energy (bbl/y),
and n is its operating life (y). Thus a $100 device
that saved 1 bbl/y and lasted 20 y would have a
CSE of $8/bbl. Against a $26/bbl oil price, a 20-y
device with a 1-y simple payback would have a
CSE of $2.1/bbl. Engineering-oriented analysts
conventionally represent efficiency “resources”
as a supply curve, rather than as shifts along a
demand curve (economists’ convention). CSE is
methodologically equivalent to the cost of
supplied energy, such as refined petroleum
217. Small & Van Dender 2004. The authors’ findings, used here by kind
permission, should not be ascribed to their research sponsors (nor
should our findings). Their calculated long-run elasticity should
decline further with increasing income. Even from an economic welfare perspective, net rebound effects are small enough to have been
neglected by NAS/NRC 2001a, p. 89.
218. Details are in Technical Annex, Ch. 4. We count Intelligent
Highway Systems here rather than in the supply curve of oil-saving
potential because its cost is hard to determine (note 380, p. 78).
219. See next page.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
products: the value of the energy saved isn’t
part of the CSE calculation, which shows only
the cost of achieving the saving. Whether that
saving is cost-effective depends on whether its
CSE is more or less than the cost of the energy
it saves, and on what costs are counted (private
internal vs. full societal). Except as noted, our
CSEs are compared on the short-run margin
with EIA’s RAC of $26.08 (2000 $) in 2025, excluding externalities and downstream capital costs.
CSEs can be negative if capital charges are
more than offset by saved maintenance cost,
avoided equipment, etc.
Rebound: People may drive more miles in moreefficient light vehicles because their fuel cost
per mile drops. Although their decision to do so
indicates that they value the increased mobility
more than the cost of the fuel consumed, they
will use more fuel, and we’re calculating oil use,
not economic welfare. Older estimates of ~10%
rebound, or even up to 30% long-term, now
appear overstated, and are about halved by the
latest econometric evidence.217 Variabilized
insurance payment (p. 218) and the need to
replace lost gasoline tax revenues by equivalent
user fees (pp. 212–213) offset 73% of the ¢/mile
reduction from 2025 State of the Art ’s 69% fuel
savings. This reduces the long-run driving
rebound to 3.3%. Intelligent Highway Systems
(p. 78) then save 1.8% of light-vehicle fuel, leaving a 1.5% net rebound.218 We’re comfortable
neglecting that because of a larger unanalyzed
issue with EIA’s Reference Case: it assumes
each driver travels 36% more miles in 2025 than
in 2000, yet that extra driving time, exacerbated
by congestion, will collide with other priorities
in people’s already-saturated time budgets—
and with their 64% higher per-capita disposable
incomes, they’ll value time even more highly
than they do today.219
(continued on next page)
41
This Report
Box 5: Conventions (continued)
Interaction between efficiency improvements:
Successive energy savings don’t add; they multiply. That’s because each leaves less energy
use to be saved by the next. We count this effect
fully. Some savings incur additional costs or,
more often, synergistic benefits: e.g., better
aerodynamics or tires can reduce the weight
and size of a vehicle’s engine. We count such
effects carefully for light vehicles, by relying on
an archetypical design specifically optimized for
that purpose, and partially (tending to understate
oil savings) for heavy trucks and for aircraft.
219. Like EIA, we have not explicitly analyzed rebound for trucks or
aircraft, both because they’re much smaller oil-users and because of
the lack of literature. However, trucking demand tends to be driven by
industrial production, which doesn’t go up just because trucks
become more efficient. Airlines can rarely pass through higher fuel
costs today, but if competition did allow them to pass through future
fuel savings, that might increase flying somewhat. Meanwhile, however, there will be increased competition from efficient high-speed rail,
videoconferencing, etc.; and in our view, travel hassle is likely to constrain the already-high EIA forecasts of air travel from rising further.
42
Turnover of capital stocks: The following analysis first presents the technical potential as if
all the efficiency improvements and supply
substitutions shown were (by magic) fully adopted by 2025. This helps readers to understand the
size of the “efficiency resource” independent of
how quickly it can actually be captured. Later,
on pp. 178–212, we contrast actual adoption
trajectories if we passively wait for capital
stocks to turn over normally with those achievable under innovative policies that accelerate
turnover. Thus turnover rates aren’t fixed; they’re
a policy variable. See Technical Annex, Ch. 21,
for a fuller discussion of stock turnover and
of related issues about the degree of adoption.
The simulated effects of our proposed policies
to accelerate adoption of advanced technology
vehicles are summarized graphically on pp.
180–181.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil
American oil savings were a gusher in 1977–85, but slowed to a
trickle in the mid-1980s when we closed the main valve—light-vehicle
efficiency. During 1975–2003, U.S. primary energy consumption per
dollar of real GDP fell by 43% 220—in effect, creating the nation’s biggest
energy “source,” now providing two-fifths of U.S. energy services, and
equivalent to 1.9 times 2003 U.S. oil consumption, 5.1 times oil production, 3.4 times net oil imports, and 13.9 times Persian Gulf net imports.
Per-capita U.S. primary energy use rose 0.6% while per-capita GDP
grew 78%. And we’ve saved oil even faster than total energy: the black
line in Fig. 8 shows that in 2003, producing a dollar of GDP took only
half as much oil as it took in 1975. Yet that halved intensity was achieved
despite a severe handicap: the aqua line shows that new light vehicles
(which use a third of U.S. oil) have generally been getting less efficient
for the past 23 years. The nation’s oil intensity fell anyway, by 1.8% per
year, during 1986–2003 as other sectors’ efficiency gains offset light vehicles’ efficiency losses (some shifts in the composition of economic output
may have helped too). In short, after 1985, the pace of saving oil per dollar
of GDP fell by two-thirds—yet the tripled speed of 1977–85, when we were
paying attention, could be regained if light vehicles simply resumed the
sort of rapid technological progress they were achieving two decades ago.
EIA projects, as the black line shows, that oil intensity will fall by a further 26% during 2003–25—falling only half as quickly in the next 22 years
Figure 8: U.S. oil intensity (1975–2025)
Reductions in oil consumption (black) per dollar of real GDP, and of adjusted EPA gallons per mile221 (aqua)
for new light vehicles, all compared with 1975 levels. This graph assumes as a placeholder that the oil-saving efficiency techniques this study shows to be cost-effective will be fully implemented at a linear rate
during 2005–2025 (cf. actual adoption rates simulated on pp. 180–181).
Government forecasts
assume that oil use
per dollar of GDP will
fall only half as fast
in the next 22 years
as it did in the past
15, and that cars and
light trucks take nine
years to regain their
1987 mpg. But continuing the sedate oilsaving pace of the
past 15 years, and
improving light vehicles only five-eighths
as fast as we did in
1975–81, would save
half of 2025 oil use.
Option 1.
Full use of costeffective, established
technologies can
wring twice as much
work from each
barrel by 2025 as the
government projects,
at half the cost of
buying that barrel.
Most of the savings,
like most of the use,
is in light trucks,
heavy trucks, cars,
and airplanes.
1.2
actual
1.0
actual and EIA projected
barrels oil/$ real GDP
projected
0.8
RMI 2004 State of the Art
end-use efficiency,
hypothetical full deployment
0.6
2025
2020
2015
2010
2005
2000
0
1995
RMI State of the Art
new light vehicles’ gal/mi
1990
0.2
1985
actual and EIA projected
new light vehicles’ gal/mi
1980
0.4
1975
intensity vs. 1975
oil/$ GDP –5.2%/y
year
Source: RMI analysis based on EIA 2003c; EIA 2004; ORNL 2003; EPA 2003; and this report’s State of the Art analysis.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
220–221. See next page.
43
Saving Oil
Option 1. Efficient use of oil
After 1985, the pace of
saving oil per dollar of
GDP fell by twothirds—yet the tripled
speed of 1977–85,
when we were paying
attention, could be
regained if light vehicles simply resumed
the sort of rapid technological progress
they were achieving
two decades ago.
as it actually fell in the past 15 years, a period of moderate prices and
stagnant policy. We’ll show that cost-effectively efficient use of oil, using
State of the Art technologies, could double this to another 50% cut (the
same percentage as 1975–2003’s), as illustrated by the dashed black line.
The most important change is in light vehicles. EIA assumes their gallons
per mile will fall only 9% by 2025, surpassing by only 0.5 mpg in the next
22 years the efficiency they enjoyed in 1987. In contrast, we’ll find a
potential drop not of 9% but of 72% in gallons per mile (to 73 mpg), as
shown in the dashed aqua line.
Saving oil must focus
on transportation,
projected to use
74% of oil in 2025,
and (as we’ll see starting on p. 169) on how
to reward people for
buying efficient and
scrapping inefficient
vehicles.
Cars and light trucks,
projected to burn
46% of U.S. oil
in 2025, can save over
two-thirds of their
fuel by artfully combining today’s best
techniques—without
compromise, with
better safety and pep,
at attractive cost,
and with competitive
advantage.
Fig. 8 illustrates that if, hypothetically, both these improvements were
made at a constant rate during 2005–25, they’d still both be much slower
than the steep gains actually made in 1977–85: the overall drop in oil
intensity would simply continue the sedate slope of the 1990s.
Nonetheless, the dramatic savings we propose may at first sight surprise
some readers. To understand why they’re both practical and profitable,
we must delve more deeply into each end-use of U.S. oil, starting with
transportation—which uses 27% of the nation’s energy but 70% of its oil.
Transportation
Highway transportation is a gigantic industry for which Americans pay
trillions of dollars a year (mainly unmonetized222). Its oil use is disproportionately bigger still. As shown in Figs. 6–7, in 2000, of the 70% of U.S. oil
that fueled transportation, three-fourths fueled road vehicles. Light trucks
cause 55% of the total growth in oil consumption to 2025 (Fig. 7)—3.8
times the growth share of the runner-up, heavy trucks, both distantly followed by aircraft and automobiles. We therefore emphasize oil-saving
opportunities in these four uses, especially light and heavy trucks.
Light vehicles
Every two seconds, American automakers produce a new light vehicle—
a marvel of engineering, manufacturing skill, business coordination, and
economy, costing less per pound than a McDonald’s quarter-pound ham220. The drop in intensity for primary energy used directly, not in power plants, was 52%. (Total gas intensity fell 54%; direct
[non-electric] gas intensity fell 57%.) The 43% drop in total primary energy intensity is all the more impressive because by
2003, generating electricity used 39% of all primary energy consumption, up from 28% in 1975, and electric intensity fell by
only 12% since 1975. (The modest saving, despite electricity’s being the costliest form of energy, is not surprising since
electricity is often priced at historic average cost; electricity is the most heavily subsidized form of energy; and importantly,
48 states reward distribution utilities for selling more electricity but penalize them for cutting customers’ bills—see pp.
219–220) Indeed, 44% of all growth in primary energy consumption during 1975–2003 went to losses in generating and delivering electricity. (Fortunately, those processes also became 12% more efficient.)
221. Departing from this report’s normal convention, the gal/mi values in Fig. 8 are in “laboratory” terms to conform to the
EIA projections shown, but it doesn’t matter because all values are indexed to 1975.
222. Delucchi 1998. Table 1-10 summarizes social costs of $2.0–3.9 trillion/y in 2000 $, less than half of it produced and priced
in the private sector. Vehicle-miles have grown by more than 30% since his base year. One indicator of social cost is that
U.S. passenger vehicles emit as much CO2 as all of Japan, the world’s second-largest market economy.
44
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles
burger. The trillion-dollar global auto industry is the largest and most complex undertaking in the history of the world. It meets conflicting requirements with remarkable skill. As a classic mature industry,223 it is starting to
undergo fundamental innovation that will redefine the core of how vehicles
are designed and built. That innovation is focusing on new ways to reconcile customer requirements (occupant safety, driving experience, functionality, durability, sticker price, total cost of ownership, and esthetics—cars are
vehicles for emotions as well as for bodies) with such public concerns as
third-party safety, fuel efficiency, carbon and smog-forming emissions, recyclability, fuel diversity, competitiveness, and choice.
Both Detroit and Washington have long assumed, from economic theory
and incremental engineering tweaks, that fuel-thrifty vehicles must be
unsafe, sluggish, squinchy, or unaffordable, so customers wouldn’t buy
them without government inducement or mandate. Congress has deadlocked for two decades on whether such intervention should use higher
gasoline taxes or stiffer fuel-economy standards—notably the CAFE standards signed into law by President Ford in 1975 with effect from 1978, followed by analogous Department of Transportation light-truck standards
effective in 1985, and widely believed to account for most of the dramatic
light-vehicle fuel savings shown in Fig. 8.224 Europe and Japan attained
similar or better mpg levels (typically with smaller vehicles) via high
gasoline taxes, but now find those insufficient. They and soon Canada are
implementing further 25% savings by policy.225 China’s comparable or
stiffer efficiency standards apply to every new car sold from July 2005,
and with anticipated extensions, should save 10.7 billion barrels by
2030,226 rivaling the oil reserves of Oman plus Angola. Most U.S. SUVs
would flunk China’s 2009 standards.227 Given China’s focus on building a
huge auto industry as a pillar of industrial strategy (p. 167), even more
dramatic efficiency or fuel leapfrogs will be needed for China to avoid fullfledged U.S.-style oil dependence, which could “undercut all of today’s
costly efforts by the U.S. to reform and stabilize the Middle East.” 228
Growing evidence suggests that besides fuel taxes and efficiency regulations, there’s an even better way: light vehicles can become very efficient
through breakthrough engineering that doesn’t compromise safety, size,
Most U.S. SUVs
would flunk China’s
2009 efficiency
standards.
223. Characterized by convergent products, fighting
for shares of saturated and
oversegmented core markets, at cut-throat commodity prices, with generally
low returns and global
overcapacity (by about
one-third). For U.S.
automakers, innovation of a
fundamental rather than
incremental nature was
also stagnant until the past
decade.
224. Greene 1990; although
we find this paper compelling, diverse views are
cited in both the 2001 and
the 1992 NAS/NRC reports.
Greene (1997) summarizes
the arguments.
225. The European Auto Manufacturers’ Association (ACEA) has voluntarily agreed to reduce new cars’ fuel use by 25% by 2008. The International Energy
Agency (2001) judged this feasible at low cost, although execution may be lagging. DaimlerChrysler extrapolates the improvement from an average of ~42
mpg in 2008 to 47 mpg in 2012 (Herrmann 2003, p. 2). Japan’s “Top Runner” program requires all new vehicles over time to approach the efficiency of the best
in each of eight weight classes (with some 50%-discounted trading allowed for over-/underperformance between classes); this has improved the overall
fleet’s fuel economy by about 1% a year even though vehicles have become larger (ECCJ, undated).
226. D. Ogden (Energy Foundation), personal communication, 7 April 2004. Full implementation of the new standards is expected to save a cumulative 1.6 billion bbl by 2030; with anticipated tightening, 4.8; and with expected new standards on light- and heavy-duty trucks and motorcycles, an additional 5.9, for a
total of 10.7 billion barrels. The short-term reduction in fuel intensity is roughly 15%.
227. An et al. 2003; Bradsher 2003.
228. Luft 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
229–43. See Box 6 on p. 46.
45
Saving Oil
Option 1. Efficient use of oil: Transportation
Light vehicles can
become very efficient
through breakthrough
engineering that
doesn’t compromise
safety, size,
performance, cost,
or comfort, but
enhances them all.
performance, cost, or comfort, but enhances them all. Disruptive technology could make government intervention, though potentially still very
helpful, at least less vital: customers would want such vehicles because
they’re better, not because they’re efficient, much as people buy digital
media instead of vinyl phonograph records. Automakers could then rely
on traditional and robust business models based solely on competitive
advantage in manufacturing and value to the customer, and have to
worry much less about such random but potentially harsh variables as oil
price, climate-change concerns, and elections.
Light vehicles
We first present the traditional, incremental, policy-based approach to
light-vehicle efficiency, adopting a widely respected industry analysis as
the basis for Conventional Wisdom (p. 69). State of the Art, in contrast, uses
the integrative designs and advanced technologies illustrated by some
recently developed concept and market cars. Later, when discussing public policy, we’ll analyze innovative ways to accelerate market adoption.
To ground readers’ understanding of where better car efficiency can come
from, we first offer a short tutorial on the physics of light vehicles (Box 6).
6: How do light vehicles use fuel, and how can they save fuel?
A typical recent-year production car gets about
28 EPA adjusted mpg, or 8.4 liters of fuel per 100
km, on level city streets. (To convert between
miles per U.S. gallon [mpg] and L/100 km, divide
235.2 by the other.) Where does the energy in
that fuel go? About 85–87% is lost as heat and
noise in the powertrain—engine, pollution
controls, the mechanical drivetrain transmitting
torque to the wheels—or in idling at 0 mpg
(which wastes 17%). Only ~17% reaches the
wheels in EPA testing, or ~12–13% in actual
driving using accessories; ~6% accelerates
the car. Since ~95% of the mass moved by that
229. Sovran & Blaser (2003) show that for a typical contemporary midsize
car (Table 6), tank-to-wheels efficiency is 15.5% city and 20.2% highway,
hence 17.3% combined. A standard 70-kg driver is 4.3% of the car’s test
mass, so 0.7% of the fuel moves the driver even with no idling. The average load factor of U.S. cars is not much above 1.0.
230. Sovran & Blaser (2003) give more exact values: for a typical contemporary midsize car, ~32% aero, 29% rolling, and 40% braking losses, all
sensitive to assumptions made. An & Santini (2004) give corresponding
values of 24%, 25%, and 42% for a 2000 Taurus, plus 8% accessory loads
that are not counted in CAFE testing.
46
~12–13% of the fuel energy is the car, not the
driver, the car uses less than 1% of its fuel to
move the driver—not a very gratifying outcome
from a century of devoted engineering effort.229
The ~12% (more or less in different models) of
the fuel energy that reaches the wheels meets
three kinds of “tractive load.” In round numbers,
nearly one-third heats the air that the car pushes aside (“aerodynamic drag”), one-third heats
the tires and road (“rolling resistance”), and
one-third accelerates the car, then heats the
brakes when it stops (“inertia load” or “braking
loss”).230 At low speed, the last two terms
account for 80+% of the load, but the fuel needed to overcome aerodynamic drag rises as the
cube of driving speed, doubling between 55 and
70 mph. Energy used to heat air, tires, and road
can’t be recovered, but most of the braking
energy can be.
(continued on next page)
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Box 6: How do light vehicles use fuel, and how can they save fuel? (continued)
The basic physics of these loss mechanisms is
straightforward. At a given speed and air density, aerodynamic drag depends on the product of
the vehicle’s frontal area times its drag coefficient. That number, abbreviated Cd , depends on
how far back the smooth laminar flow of air
around the car adheres before detaching into
turbulence; this reflects shape, smoothness,
and such details as wheel-well design, underbody protrusions, body seams, side-mirrors, etc.
Rolling resistance depends only slightly on
speed, but mainly on the product of the car’s
weight (how hard it presses down on the tires),
times the tires’ coefficient of rolling resistance,
r0 (which can be reduced by better tire designs
and materials, taking care not to compromise
safe handling). Both the power needed to accelerate the car and the energy dissipated in the
brakes to decelerate it rise directly with the
car’s weight. Thus two of the three tractive
loads are weight-dependent, as is any energy
used for climbing hills, so as we’ll see on p. 52,
two-thirds to three-fourths of fuel consumption
is typically weight-dependent—and we don’t
need weight for safety (see pp. 57–60).
Fuel can be saved either by needing to do less
work or by doing it more efficiently. Automakers
have traditionally focused mainly on the latter—
on tank-to-wheels efficiency, especially engine
efficiency—because that’s where most of the
losses are. (On the same logic, Willie Sutton
231. In a typical midsize car, the powertrain averages ~70% efficient
from engine output shaft to wheels (Sovran & Blaser 2003) but the
peak efficiency of an Otto engine is only about half that, and its average
efficiency under varying load is about halved again.
232. In the Lohner-Porsche Chaise, which hauled cannon for the
Austrian Imperial Army in World War I (von Frankenberg 1977, pp. 10,
17). Of course, a century ago, power electronics didn’t exist, and both
structural materials and motors were an order of magnitude inferior to
today’s. Efficient car designs have a long and fascinating history, ranging from Buckminster Fuller’s ~30-mpg, 120-mph, 11-seat Dymaxion in
1933–34 to the 1,200-lb Pertran designed by Battelle Memorial Institute
(Columbus, Ohio) in 1980 and simulated to get 80–85 mpg with an Otto or
100–105 mpg with a diesel engine.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
explained that he robbed banks because that’s
where the money is.)231 Powertrain efficiency
has risen by about one-third since 1975, though
its benefits were reversed by faster acceleration and increased weight and features (p. 8).
Further gains are still available from many
mechanical and control improvements; running
the engine in its most efficient ranges of speed
and torque; automatically turning off the engine
when idling; and cutting off fuel to cylinders
whose power isn’t needed. Such “Displacement
on Demand,” projected to reach more than two
million GM vehicles by 2008, will add ~1 mpg
to some 2004 V-8 midsize SUVs, and similarly
for Chrysler, Honda, and others.
Internal-combustion engines—both standard
Otto engines and high-compression diesel
engines, named after their German inventors—
are most efficient at high power. Yet they actually run mainly at low power, averaging about
11% of their full potential on the highway and
8% in the city, because most cars are more than
half steel, and steel is heavy. Accelerating its
weight takes so much force that the mismatch
between available and actually used engine
power cuts a typical Otto engine’s operating
efficiency about in half—worse as engines get
even bigger. Continuously variable or automated
transmissions and engine computer controls
help, but modern power electronics and
advanced electric motors permit a more fundamental solution to this mismatch.
Hybrid-electric (“hybrid”) cars, invented by
Ferdinand Porsche in 1900,232 overcome the
engine/tractive-load mismatch by turning the
wheels with various mixtures of power transmitted directly from an engine and from one or
more electric motors that decouple traction
from the engine. The electricity comes from the
(continued on next page)
47
Saving Oil
Box 6: How do light vehicles use fuel, and how can they save fuel? (continued)
Figure 9: Two mild hybrids competing with the full-hybrid
Toyota Prius in the U.S. L: 2-seat Honda Insight , 59–64 mpg.
R: The reportedly profitable 5-seat Honda Civic Hybrid,
49 mpg, ~10% share of Civic market. Insight ’s successor
might resemble the 2003 carbon-fiber concept IMAS—
698 kg, Cd 0.20, 94 mpg on the Japanese 10/15 cycle.
engine or from stored energy recovered from
the motors when the car brakes or coasts downhill; hybrids don’t plug in for recharging like battery-electric cars. Modern hybrids entered the
Japanese market with the first Toyota Prius in
1997, a decade or two earlier than the industry
expected. Honda brought hybrids to the U.S. in
1999 (Fig. 9), Toyota in 2000.
In “hybrid-assist” or “mild hybrid” designs like
these Hondas, a small electric motor boosts the
engine’s power for acceleration and hill-climbing. The small engine, run mostly at or near
its “sweet spot,” can then act like a bigger one
and boost efficiency by ~25% (~66% for Insight
including mass and drag reduction).233 In
Toyota’s full-hybrid 2004 Prius (5 seats, 55 mpg,
pp. 29–30), planetary gears split the gasoline
engine’s shaftpower between wheelpower and
generating electricity, smoothly providing a
computer-choreographed mix of mechanical
and electric drive (the latter can run the car on
its own at up to 42 mph for a quarter- to a halfmile). This more complex arrangement slightly
more than doubles efficiency: Toyota reckons
that in U.S. average-driving tests, its production
models’ overall efficiency in ton-mpg is about
18% for typical 2003 non-hybrid cars, 27% for
the 1998 Prius, 31% for the incrementally refined
2003 Prius, and 37% for the redesigned 2004
Prius, which uses the Atkinson variant of an
Otto engine better suited to hybrid operation,
augmented by electric drive nearly as powerful.234 Toyota now sells seven gasoline- and
diesel-hybrid models in Japan, from subcompacts to delivery trucks.235
The 2004 Prius is not only ~104% more efficient
in ton-mpg than a modern non-hybrid; it also
gets 38% more ton-mpg than the 1998 Prius.
That six-year gain beats the average new U.S.
light vehicle’s gain during the past 26 years
(1977–2003).236 Just from the 2003 to the 2004
Prius, Toyota reports that its permanent-magnet
motors gained 50% in power and 30% in peak
torque; its traction batteries gained 35% in
power/weight;237 its inverter became more powerful and efficient while the size of its powerswitching transistors shrank 20%; its engine got
10% more power per liter; and its overall powertrain efficiency on a ton-mpg basis rose 19%
(continued on next page)
233. An et al. 2001; DeCicco, An, & Ross 2001. The latter paper states at p. B-12 that Honda ascribes 30% of Insight’s urban-cycle mpg improvement
to mass and drag reduction, 30% to the efficient lean-burn engine, and 25% to the hybrid powertrain, but synergies between these probably raise
hybridization’s share of the mpg gain to ~40%, corresponding to ~25% higher mpg.
234. A useful technical presentation from Toyota is posted at John’s Stuff 2004. The Atkinson engine is apparently made by Nissan, to which Toyota has
licensed the Synergy Hybrid Drive’s power electronics and transmission (both technology and actual components) so that Nissan can sell 100,000
hybrids during 2006–11 (Peter 2003).
235. Fairley 2004. Models include the Estima 7-seat minivan (18.6 km/L on Japan’s 10/15 test mode) and the Alphard 8-seat 4WD minivan (17.2 km/L)
with a 1.5-kW AC power outlet. At the approximate 1.37 conversion factor from these Japanese test results to EPA adjusted mpg (ORNL 2003, Table
4.26), these models would respectively get roughly 60 and 55 EPA adjusted mpg.
236. EPA 2003, Table 1.
237. They are also designed to last the life of the vehicle, partly by controls designed to prevent deep discharge (and probably overcharge).
Next will come even lighter, cheaper, longer-lived ultracapacitors, as in Honda’s fuel-cell FCX.
48
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles
The conventional view
In July 2001,244 a panel of the National Academies’ National Research
Council (NRC) found that under pure competition not driven by regulation, new midsize cars in 2012 could improve from 27.1 to 30.0 mpg at an
extra retail price of $465, and with aggressive development of unproven
but basically sound technologies, such cars could achieve 41.3 mpg by
2012–17 at an extra price of $3,175—all assuming unchanged vehicle per244. The report was prepublished at the end of July 2001 and formally released in mid-January 2002.
Box 6: How do light vehicles use fuel, and how can they save fuel?
(continued)
(three-fifths from better software controls and
regeneration, which recovers 66% of braking
energy 238). Emissions also dropped 30% and
production cost fell. The hybrid powertrain’s
extra weight was cut to ~70–75 kg239 or ~5%,
then to ~1% (offset by reduced drag240) by
weight savings elsewhere, so contrary to industry expectations, hybrids “can provide significant improvements in fuel economy with little
or no change in [net] mass.”241 After one and a
quarter centuries of maturation, modern nonhybrid powertrains offer far less scope for
such impressive gains.
Automakers want to shift from Otto to ~25–30%more-efficient242 diesel engines, which intensive
development (mainly in Europe, where they’re
nearing 50% market share) has lately made far
cleaner and quieter than rattling and soot-belching 1970s models. Of course, Prius or any other
hybrid could also adopt small diesel engines,
A National Academy
panel’s 2001 finding
that light vehicles could
achieve major fuel
savings—without compromising size, comfort,
safety, performance,
or affordability—
is already way behind
the market.
boosting its overall efficiency in proportion to
that of the engine. Conservatively, this study
assumes no light-vehicle diesel engines (p. 71).
Even more fundamental efficiency gains can
come from making cars lighter in weight (but
also safer, thanks to advanced materials and
designs—pp. 55–71) and lower in aerodynamic
drag and rolling resistance. These “platform
physics” advances, keys to State of the Art
designs (pp. 62–71), can then be valuably combined with hybrid powertrains and other propulsion improvements, gaining a more-than-additive
(synergistic) mpg advantage 243 and manufacturing cost reduction.
Much smaller but still useful fuel savings can
also come from more efficient accessories, such
as metal-halide headlights, visually clear but
heat-blocking windows, and more efficient airconditioners and fans. These are applicable to
all kinds of vehicles, and become more important
as savings in propulsion fuel make accessory
loads a larger fraction of the remaining fuel use.
238. An & Santini 2004. Even better future regenerative braking recovery could even make the vehicle more efficient than its engine (An & Santini
2004): the 2004 Prius and its engine are 35% efficient, while the theoretical SAE limit on an Otto engine’s peak efficiency is 38%.
239. D. Hermance (Executive Engineer, Toyota USA), personal communication, 6 February 2004.
240. None of the platform’s efficiency gain is attributed to improved physics (weight, aerodynamic drag, and rolling resistance), because the better
drag coefficient and any tire improvements were offset by increases in weight and perhaps in frontal area.
241. An & Santini 2004.
242. The efficiency gain doesn’t include diesel fuel’s advantage of 15% more miles per gallon or 13% fewer gallons per mile than gasoline even if
engine efficiencies are identical, simply because it’s a 15% heavier distilled product. Care must be taken in comparing the efficiencies of diesel- and
gasoline-fueled vehicles to ensure that the fuel difference has been properly adjusted for and the convention described, e.g., note 349, p. 71.
243. An & Santini 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
49
Saving Oil
245. The federal/Big Three
collaboration called the
Partnership for a New
Generation of Vehicles
developed these tripled-efficiency midsize concept
sedans and much significant enabling and manufacturing technology, and
established valuable collaborations that moved government technology into the
private sector. Its successor program, FreedomCAR,
has less clear objectives
and status (Lovins 2002).
Option 1. Efficient use of oil: Transportation: Light vehicles: The conventional view
Figure 10: Three 2000 PNGV245 diesel-hybrid midsize concept sedans and their gasoline-equivalent mpg
L to R: GM Precept (1,176 kg, Cd 0.163, 80 mpg),246 Ford Prodigy (1,083 kg, Cd 0.20, 70 mpg),247
and Dodge ESX3 (polymer body, 1,021 kg, Cd 0.22, 72 mpg). Of their efficiency gains, totaling 2.7–3.1× vs.
the 26-mpg Taurus-class base vehicle, ~15–18% came from switching to diesel engines, 36–47% from
smaller loads and engines (better platform physics, accessories, and their consequences), and 43–48%
from hybridization.248 ESX3 ’s cost premium over a Chrysler Concorde at 80,000 units/y was cut from
the initial 1996 ESX ’s $60,000 to $7,500, equivalent to a ~$28,500 MSRP.249 However, none of these concept
cars has yet been turned into a market platform due to their higher production cost, which is ascribed
largely to costlier light metals and greater powertrain complexity.
246. Dunne 2001.
247. Edmunds.com 1999;
Electrifyingtimes.com 2000;
Moore 2000.
248. NAS/NRC 2002a
at Fig. 3-14.
249. Manufacturer’s
Suggested Retail Price.
McCraw 2000; Automotive
Intelligence, undated.
250. NAS/NRC 2002, p. 45.
251. An et al. 2001.
252. NAS/NRC 2002, p. 53.
253. NAS/NRC 2002, p. 40.
254. U.S. hybrid sales
totaled about 38,000 in 2002
and 54,000 in 2003, according to J.D. Powers and
Associates. Although that’s
still modest in the United
States’ 16.7-million-lightvehicle 2003 market, hybrid
sales grew by nearly 89%/y
during 2000–03 according
to R.L. Polk & Co. data
(Wall St. J. 2004). J.D.
Powers predicts sales of
about 107,000 in 2004 and
211,000 in 2005. However,
Toyota is said to project its
own U.S. hybrid sales alone
at 300,000 in 2005, and at
May 2004 had a 6–12-month
backlog of Prius orders
(Silverstein 2004).
formance and a 3.5% efficiency penalty from heavier safety systems.250
Oddly, the panelists excluded hybrid propulsion. They noted the Big
Three’s diesel-hybrid concept family sedans251 rated at 70–80 mpg (Fig.
10), but concluded only that such advanced hybrids “would cost much
more than conventional vehicles”:252 hybrids, they found, might boost
fuel efficiency by 15–30% at an extra cost of $3,000–7,500+ if more than
100,000/year were made, but weren’t yet realistic. Global sales of 85,000+
hybrid cars during 1997–2001 didn’t make hybrid propulsion listable even
among “emerging” technologies in 2001, let alone those with “production
intent.”253 (Through 2003, more than 200,000 were sold.254)
Applying traditional, modest, incremental improvements, including only
minor reductions in weight and drag,255 the panel found that mpg gains of
14–53%256 (if weighted by the 2002 sales mix 257) would raise prices by
~$168–217/mpg—a 6–8-y simple payback at ~$1.50/gal. Applying the
larger savings instantaneously to the 2001 fleet would reduce U.S. oil use
by 2.7 Mbbl/d,258 more than imports from the Gulf. Thus the 2001 NRC
study found a technical potential for major fuel savings without compromising safety, performance, or affordability. This
…would necessitate the introduction of emerging technologies, which have the
potential for major market penetration within 10 to 15 years. These emerging
technologies require further development in critical aspects of the total system
prior to commercial introduction. However, their thermodynamic, mechanical,
electrical, and controls features are considered fundamentally sound.
255. While correcting some findings in its initial report, in NAS/NRC 2002, the panel later correctly noted that it “may have underestimated” the near-term
(10–15-year) savings available from reducing vehicles’ mass and drag (Wall St. J. 2002). Its revision made the savings bigger and cheaper.
256. Based on the revised 2002 data (NAS/NRC 2002, p. 18) the potential mpg gains in that scenario were 11–43% for small, 11–52% for midsize, and 12–58%
for large cars; 11–51% for small, 20–63% for midsize, and 20–65% for large SUVs; 15–59% for minivans; and 17–58% for small and 15–59% for large pickup
trucks, all at increased retail prices ranging from about one to five thousand dollars. See Fig. 11.
257. The latest stable data available at this writing: ORNL 2003, p. 4-9.
258. Applying a 35% improvement to the actual 2002 fleet’s on-road efficiency of 20.2 mpg (ORNL 2003), its 2.56 trillion vehicle-miles, and a 0.94 multiplier to
convert energy content per barrel from gasoline to crude oil, and assuming the panel’s “Path 3” (highest-implementation) scenario.
50
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: The conventional view
These 2001 findings continued two
historic trends (Fig. 11): savings become bigger and cheaper over time,
and the NRC panel was fully as
conservative as one might expect.259
Figure 11: 1990–2004 comparison of absolute mpg vs. incremental costs
for new U.S. automobiles
Smoothed automotive-efficiency supply curves from the 1992 and 2001 National
Academies’ National Research Council analyses, compared with a credible line
of independent analysis they declined to adopt. Both sets of studies trend
toward the lower right over time, and all have turned out empirically to be
conservative, as shown by the 1992 Honda VX subcompact and the 2004 Prius
hybrid midsize sedan (both in red).
60
55
50
45
40
35
30
25
price increase (MSRP 2000 $)
Some found the NRC’s 2001 findings implausibly optimistic, and a
5,000
2002 revision made them a tad
4,500
2004 Prius
NRC
more so.260 But the NRC seriously
(2004 actual to
4,000
High
~2007
goal)
2001
underestimated the pace of pri3,500
DeCicco
Ledbetter
vate-sector innovation. Just 34
3,000
& Ross
& Ross
NRC Low
NRC
1990 Avg 2001
weeks after its report was released,
1995
2,500
High
Full Avg
1992
the 2004 Prius came on the market.
2,000
NRC Low
It beat the NRC’s forecast for mid1,500
DeCicco, An,
1992
Ross Mod & Adv 2001
size sedans (27.1 mpg) by not the
1,000
1992 VX subcompact
projected 20 but 28 mpg, not in
500
2012–17 but in 2003, at the predict0
ed price increase of $2,000–4,000
but largely shaved from the manuabsolute miles per U.S. gallon
(EPA adjusted, combined city/highway)
facturer’s margin, not added to the
Source: RMI analysis, see References for sources.
sticker price.261 Nor was this the
NRC’s first such embarrassment
by the market: its 1992 auto study was published only weeks before the
51-mpg 1992 Honda VX subcompact hatchback entered the U.S. market.
259. The American Council
It was 16% more efficient, and its 56%-in-one-year mpg gain was 32–73%
for an Energy-Efficiency
cheaper, than the NRC had deemed feasible with “lower confidence” for
Economy (ACEEE)
advanced-technology
14 years later (model year 2006).262 In a 1991 symposium to inform that
analyses, graphed for comNRC study, the Big Three had claimed that cars could be made only
parison, were mentioned
by the panel as assuming
about 10% more efficient without making them undriveable or unmarmuch more optimistic techketable—a vision framed by dividing the future into “too soon to change
niques than the panel considered, but their substance
anything” and “too far off to speculate about,” with nothing in between.
was rejected without explaFortunately, Honda in 1992 and Toyota in 2003 felt uninhibited by this
nation. Apparently the panel
was simply unable to deal
cramped vision, and market share now rewards their boldness. But even
with this serious, independmore fundamental efficiency advances are now emerging within the
ent, but for some members
uncomfortable, analysis—
industry—innovations that can shift the curves in Fig. 11 even further
which, as we’ll see, also
toward the lower right.
proved conservative.
260. NAS/NRC 2002; these revisions made the 2001 preliminary edition’s predicted savings slightly cheaper, particularly for cars.
The critiques include Sovran & Blaser 2003 and Patton et al. 2002.
261. Public data do not permit a comparison of possible differences in manufacturer’s margin between the 2004 Prius and such benchmark vehicles as the
2004 Camry or Corolla, but as reported in note 178, Prius is informally said to earn a ~$1,100 manufacturer’s margin. The base price (before destination
charges) of the Prius was constant in nominal dollars, hence declining in real terms, from U.S. market in 2000 until April 2004 (Toyota 2004b).
262. NAS/NRC 1992, p. 152 (projecting as “technically achievable” a 44-mpg subcompact at a marginal retail price of $1,000–2,500 in 1990 $; average MY1990
subcompacts were EPA-rated 31.4 mpg); cf. Koomey, Schechter, & Gordon 1993. They found the 1992 U.S. (not CA) version VX ’s adjusted EPA city/
highway rating of 50.9 mpg incurred a marginal retail price, adjusted for other model differences, of $726 in 1992 $ (equivalent to $684 in 1990 $ or $845 in 2000 $),
consistent with engineering estimates of the added production cost. This retail price was equivalent, at a 7%/y real discount rate, to $0.77/gal in 1992 $,
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
51
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles
The key to the next
efficiency breakthrough, previously
slighted, is ultralight
materials now
entering the market.
Advanced automotive technologies:
lightweight, low-drag, highly integrated
Weight causes twothirds to three-fourths
of total fuel consumption. It’s more important to make a car
light and low-drag
than to make its
engine more efficient
or to change its fuel.
Much of the fuel
used to overcome air
and rolling resistance
can be routinely
saved at low or no
cost by careful
engineering design.
To catch up with and extend modern technology requires a deeper and
wider view of what’s possible, not just with hybrid and other advanced
propulsion options, but especially in reducing light vehicles’ weight and
drag. Because a typical car consumes about 7–8 units of fuel to deliver
one unit of power to the wheels, every unit of tractive load saved by reducing
weight and drag will save 7–8 units of fuel (or ~3–5 units in a hybrid).
Automakers traditionally consider compounding losses as energy flows
from tank to wheels, but it’s more fruitful to start at the wheels, save
energy there, and turn those losses around backwards into compounding
savings at the fuel tank. This requires systematic reductions in drag,
rolling resistance, and especially weight, which is causally related to
two-thirds to three-fourths of the total fuel consumption of a typical midsize
sedan.263 Contrary to folklore, it’s more important to make a car light and
low-drag than to make its engine more efficient or change its fuel. Yet this platform-physics emphasis has had less systematic attention than it deserves:
weight reductions especially have been incremental, not yet radical.
Drag and rolling resistance Low aerodynamic drag needn’t sacrifice
styling: even a brutish pickup truck can do it. The most important step is
making the underside of the vehicle (which causes about one-fourth of
the air drag) as smooth as the top, since the air doesn’t know which side
it’s on. Low drag coefficients need careful design and construction but
don’t cost much,264 and may even cut total vehicle cost by downsizing the
powertrain. Fleetwide Cd has thus fallen 2.5%/y for two decades.265
It was ~0.55 for new American cars (0.6–0.7 for station wagons) and ~0.45
for new European cars in 1975, when a distinguished group of physicists
concluded that “about 0.3–0.5 is probably near the minimum for a practical automobile….”266 Today, most production cars get ~0.3; the 2004
Toyota Prius, Mercedes C180, and Opel Calibra, 0.26; the Opel A2 1.2 TDI,
Lexus LS/AVS, and Honda Insight, 0.25. GM’s 1999 battery-electric EV1
got 0.19. GM’s 2000 Precept concept sedan (Fig. 10a) cut Cd to 0.163,
approaching Ford’s mid-1980s laboratory records (Probe IV, 0.152, and
[fn 262 cont.] or 45% less
than the 1992 average
taxed gasoline price). Less
than one-tenth of the 56 percentage points’ increase in fuel economy in the 1992 VX (vs. the 1991 DX) came from reduced weight and drag,
the rest from powertrain improvements.
263. An & Santini 2004, Table 5 gives data for a 2000 Taurus on the CAFE combined city/highway cycle augmented to break out 8% of fuel consumption due to
accessory loads; without these, 73% of fuel use would be weight-related.
264. Automakers told OTA (1995, p. 76) that “current [mid-1990s] body assembly procedures and existing tolerances were adequate to manufacture vehicles
with Cd levels of 0.25 or less,” and provided data from which OTA estimated marginal MSRP ~$138–166 (2000 $) for Cd 0.20–0.22. Low drag also raises the
fixed costs of development, but to a degree much reduced by modern computational fluid dynamics and by interaction with styling.
265. De Cicco, An, & Ross 2001, p. 10, citing 1997 Interlaboratory Working Group data.
266. American Physical Society 1975.
52
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Saving Oil
Probe V, 0.137).267 Most light trucks today are still ~0.4–0.5, due largely to
inattention, and have a huge frontal area—around 2.5–3.5 m2, vs. ~2.0 for
U.S. passenger cars accommodating the same people.268
Tiremakers have also developed much-lower-drag versions without compromising safety or handling. Compared to the mid-1970s r0 norm of
0.015 or the 2001 norm of ~0.009 269 (~0.010–0.011 for SUVs), 0.006 is “not
uncommon” for the best car tires even today,270 and 0.005, tiremakers said
a decade ago, “could be a realistic goal for a ‘normal’ [average] tire in
2015,” with some therefore even lower, at a marginal price around $130
per car.271 Test procedures and labeling requirements emerging in
California will soon let buyers discover how efficient their tires are (previously a secret). The best tires, developed for battery-electric cars, are nearly twice as efficient as normal radials of a few years ago (even with selfsealing and run-flat models that can avoid the space and weight of carrying a spare tire and jack); they could save around a tenth of fuel use in a
typical sedan. Rolling resistance also declines as the square root of tire
pressure. And lighter wheels and tires cut rotating inertia, which is equivalent to a ~2–3% heavier car.272
Lightweighting: the emerging revolution Lightweighting cars is more
controversial because it can affect both cost and safety—but not in the
way often assumed: both can now improve, markedly and simultaneously.
The 2001 NRC graph of gallons per 100 ton-miles vs. weight (Fig. 12)273
shows diverse vehicles, spanning about a threefold range in both weight
and efficiency. The key point is their vertical scatter: at a given weight they
varied in efficiency by typically ~1.6× and up to ~2.3×. This suggests, consistent with the NRC findings, that fleet efficiency could be about doubled
by adopting in all cars the best powertrain and drag-reducing designs now
used in some. But that doesn’t count potential weight reduction, to which
Taking as much as
another ton out of our
vehicles was long
assumed to be unsafe
and unaffordable.
Today, lightweighting
needn’t be either, thanks
to advances in both
metals and plastics.
267. Today’s best U.S. midsize and large production sedans only narrowly beat the 0.28 Cd measured by VW for a streamlined 1921 car, the Rumpler
Tropfenwagen, and no current production vehicle yet approaches the 0.21 Cd achieved in 1935 by the Tatra 77a (www.team.net/www/ktud/
Tatra_history_auto3.html). Perhaps the best value recorded for any enclosed vehicle—for a Swiss recumbent one-person solar-powered racecar, the 1993
Spirit of Biel III—is 0.088, falling to 0.05 if the airstream is 20˚ off-axis—but it’s doubtful a real car can achieve this even with huge hypothetical advances in
boundary-layer control. The theoretical minimum, for a perfect teardrop ~2.5–4 times as a long as its maximum diameter, is 0.03–0.04.
268. The ~2.3 m2 effective [aerodynamic] frontal area typical of new U.S. cars compares with 1.9 m2 in the 2003 Honda Insight and Civic, 1.71 in the 1991 GM
Ultralite, 1.64 in the 1987 Renault Vesta II, and perhaps ~1.5 possible with the rearranged packaging permitted by fuel-cell hybrids. But those figures are all
for the compact size class, and people are not getting any smaller—rather the opposite. Crossover designs, providing SUV-and-sedan functionality but built
like a unibody car, not a framerail truck, may help to reduce the clearly excessive frontal areas of SUVs while retaining their other customer attributes. For
example, Hypercar’s Revolution crossover design, with the functionality of a midsize SUV, has A 2.38 m2, vs. 2.90 for a typical midsize SUV. It’s 10% lower and
6% shorter than a 2000 Ford Explorer, but has equal or greater passenger headroom and legroom due to its better packaging.
269. Weiss et al. 2003, p. 22.
270. Sovran & Blaser 2003.
271. OTA 1995, p. 78. OTA also noted that brake drag and bearing-and-seal drag in the mid-1990s caused ~6% and ~12% of rolling resistance, respectively.
The former can be virtually eliminated (as in motorcycles) and the latter much reduced, supporting reductions of rolling resistance totaling 25% in 2005 and
≤40% in 2015 (OTA 1995, p. 79).
272. Sovran & Blaser 2003.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
273. NAS/NRC 2002.
53
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Figure 12: Weight vs. energy intensity (gallons of fuel needed to drive 2,000 lb of vehicle curb weight 100 miles);
each point is one vehicle model in the MY2000–01 U.S. new-light-vehicle fleet. The highway cycle test average is 1.68.
2.8
2.4
2.0
1.6
1.2
6800
6400
6000
5600
5200
4800
4400
4000
3600
3200
2800
2400
0.4
2000
0.8
1600
gallons/ton of vehicle weight–
100 miles
3.2
curb weight (lb.)
Source: NAS/NRC 2001.
Extra-strong steel
alloys and innovative
structures could
double automotive
fuel economy and
improve safety at no
extra cost.
the panel’s report gave only two inconclusive sentences.274 The panel had
been offered data on the potential of ultralight body materials, advanced
aerodynamics, and fuel-cell propulsion, but it declined to consider them,
in the unexamined beliefs that none could matter within a decade and
that lightweighting would unduly compromise both cost and safety.
They had a point about cost. Most automakers’ lightweighting experience
was then limited to aluminum and magnesium. Saving a pound of weight
via these light metals costs ~$1–3, but saves only about a gallon of gasoline every 12 years,275 so it has long been barely cost-effective against U.S.
gasoline, though it’s widely done in Europe and Japan where taxes make
gasoline prices ~2–4× higher.276 European automakers typically tolerate
weight savings costing up to €5/kg (~$3/lb) in light vehicles, and up to
six times that in some truck applications.277 In the mid-1980s, many
European and Japanese automakers even built light-metal-dominated
4–5-seat concept cars weighing as little as 1,000 pounds, using internalcombustion engines but with 2–4× normal overall efficiencies.278
274. NAS/NRC 2002, p. 39. The NRC’s analysis included only trivial weight and drag reductions of a few percent (pp. 3-20–3-22 in the 2001 preliminary
edition), all assumed to cost more, as they would if weight were cut by the traditional means of substituting light metals for steel in a few components.
275. Greene & DeCicco (2000), quoting EEA data at p. 500, state that gal/mi in a typical 1,400-kg 1998 car is reduced by 0.54% (0.64% with engine downsizing for constant acceleration) for each 1% decrease in vehicle mass. The corresponding elasticities are –0.22 for reductions in Cd and –0.23 in rolling
resistance.
276. Many production and prototype platforms in the 1980s weighed only about half as much as the U.S. MY2003 compact-car average of 1,430 kg: e.g.,
VW’s 5-seat Auto 2000 (779 kg), Peugeot’s 5-seat 205XL (767 kg), Volvo’s 4-seat LCP 2000 (707 kg), VW’s 4-seat E80 diesel (699 kg), British Leyland’s 4-seat
ECV-3 (664 kg), Toyota’s 5-seat AXV diesel (649 kg), Renault’s 4-seat Vesta II (475 kg), and Peugeot’s 4-seat Eco 2000 (449 kg), among others, had all been
reported by 1988. See Bleviss 1988.
277. COMPOSIT, undated.
278. Bleviss 1988; examples in note 276.
54
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Most looked rather costly; none came to market. Since then, light-metal
manufacturing has become cheaper, so it’s creeping into cars part by
part.279 New processes might even halve the cost of superstrong titanium,280 bringing it nearer practicality. But the belief that lightweight
necessarily means light metals and hence high cost is deeply embedded.
Now that belief is unraveling at both ends: steels are getting stronger and
light metals unnecessary. First, driven by competition from light materials
and by automakers’ need for light, strong, more formable, lower-cost
materials, steelmakers continue to develop new products: half the steel
alloys in today’s GM vehicles didn’t exist a decade ago. In 2002, a global
consortium of 33 steel companies reported an ULSAB-AVC design showing that extra-strong steel alloys and innovative structures could double
automotive fuel economy and improve safety at no extra cost.281
Saving Oil
Lightweighting no
longer requires costly
light metals, thanks
to advances in
lightweight steels
and advanced
polymer composites.
Meanwhile, an even bigger gamechanger is emerging—composites whose
polymer resins (solidified by heat, chemical reaction, or other means)
bind embedded glass or other strong reinforcing fibers, of which the most
Figure 13: Four composite concept cars: From L–R, they are: Daihatsu 2003 2+(2)-seat UFE-II hybrid (569 kg, carbon-fiber, Cd 0.19, 141
mpg on Japanese 10/15 cycle, by-wire); 282 1996 4-seat Coupé (1,080 kg including 320 kg/25 kWh NaNiCl batteries, pure-electric, 100-mi
range with a/c on, 12–20 DC kWh/100 km, 114–190 mpg-equivalent) developed by Horlacher in Switzerland for Pantila in Thailand;
BMW 1999 Z22 (~20 body parts, ~1,100 kg [~30% weight cut] via carbon and other composites and light metals, 39 mpg, by-wire);283
VW 2001 “Ein-Liter-Auto” 2-seat tandem 1-cylinder diesel 284 (carbon fiber, 290 kg, Cd 0.159, 8.5 hp, 74 mph, 238 mpg).285 (A 1990
carbon/aramid 2-seat Swiss electric car weighed just 230 kg without batteries.)286
279. Sometimes this shifts an automaker’s entire line, as with Nissan, which plans to cut all models’ weight by an average of 5–10% in the next five years to
tap demand for fuel efficiency (Kyodo News 2004).
280. This corrosion-resistant aerospace metal has half the weight, twice the strength, and currently 2–10 times the cost per part of steel. Nonetheless, it’s
starting to find selected applications, including saving ~180 pounds in the springs in the 2001 VW Lupo FSI (Das 2004).
281. The UltraLight Steel Auto Body-Advanced Vehicle Concept analysis assumes a high production volume of 225,000/y. The 5-seat midsize sedan’s virtual
design, with Cd 0.25 and curb mass (mc) 998 kg (gasoline) or 1031 kg (diesel), was predicted to achieve respective EPA combined city/highway ratings of
52 and 68 mpg with respective production costs (apparently in 2001 $) of $9,538 and $10,328. The body would weigh 52 kg less than normal, cost $7 less,
and have 81 rather than 135 parts: American Iron and Steel Institute 2002; Porsche Engineering 2002.
282. Daihatsu Motor Co. 2003; see also Motor Trend ’s impressions at Motor Trend 2003. The minihybrid uses a 0.66-L 3-cylinder gasoline Atkinson engine
buffered by NiMH batteries.
283. Birch 2000; Ganesh 2000.
284. Schindler 2002; Warrings 2003; Volkswagen, undated. One source (Endreß 2001) describes this concept car as the basis for a future market platform
cheaper than the ~DM27,000, 78-mpg Lupo, but the authors understand that VW has indicated to the contrary.
285. The literature is unclear about whether the diesel vehicles’ mpg ratings have been adjusted to gasoline-equivalent terms as the PNGV platforms in Fig. 8
were (using their 1.11 ratio of Lower Heating Value for the respective fuels). The Daihatsu’s claimed rating is on the Japanese 10/15 cycle, and the VW 2seater’s to a European test mode, neither of which is directly comparable to the U.S. rating system. A 2001 Argonne National Laboratory estimate (ORNL
2003, p. 4-32) indicates that for a typical midsize car, a Japanese 10/15 rating will be ~73% and a New European Driving Cycle rating ~92% of an adjusted EPA
rating, but that cannot be accurately extrapolated to very light and efficient cars. (If it could, the Daihatsu would get ~192 and the VW ~259 EPA mpg! These
concept cars are remarkably efficient, but probably not that efficient.)
286. The carbon-and-aramid body of this OMEKRON concept car weighed 34 kg; its 490-kg curb mass included 260 kg of batteries. The envelope was similar to
Miata’s. The design and fabrication were by Peter Kägi, then with ESORO (www.esoro.ch).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
55
Saving Oil
BMW is developing
carbon fiber for use
in series production
cars because it’s
“50% lighter than
steel” and “performs
extremely well in
vehicle crash testing.”
287. Composites World 2004a.
288. Most cost discussions
focus on cost per pound of
carbon fiber. Bulk creel
carbon of intermediate
strength and stiffness sells
in 2004 for ~$5–8/lb. This
has fallen by one or two
orders of magnitude in the
past few decades and is
coming down more with
volume and with process
innovations, such as the
~18%-cheaper microwavecarbonizing machine being
tested at Oak Ridge
National Laboratory (ORNL
2004). Price fluctuations
may also soon be damped
by proposed futures and
options markets. But at
least as important is the
major scope for reducing
the manufacturing cost of
finished composite parts.
289. Das 2004.
290. Volkswagen AG 2002.
Carbon fiber also has a
marketing value prized in
cosmetic trim (Patton 2004).
291. Brosius 2003,
pp. 32–36. Brosius 2003a
and Brosius 2003b provide
helpful background on
rapid advances in the wider
world of automotive
composites, often using
reinforcing fibers weaker
than carbon.
292. Diem et al. 2002.
293. Kochan 2003.
Meanwhile, MG is selling a
X-Power SV carbon-fiber
roadster: although made
from costly “prepreg”
carbon-fiber cloth, it has
solved several key manufacturing problems while
reportedly cutting bodypanel weight and cost
by three-fourths (JEC Composites Magazine 2003).
56
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
promising is carbon fiber. Such “advanced” (stronger than Fiberglas®)
composites, which already use more than 11 million pounds of carbon
fiber per year in worldwide sporting goods, are finally starting to transition from fancy concept cars (Figs. 13, 18) and Formula One racecars to
serious market platforms.
Carbon composites have long been used in aerospace because they’re
stronger and tougher than steel but one-fourth as dense (one-third in finished composites with 55–60% fiber content), and their strength can be
directionally oriented to match load paths and save the most weight.
Boeing plans ~25 tonnes of advanced composites, more than half the total
structural mass, in its 7E7; Airbus, ~30 tonnes in its superjumbo A380.287
The main issue is cost. Civilian aircraft can justify paying far more than
$100 to save a pound; some space missions, more than $10,000. Aerospace
composites, laid up by hand like fine couture, are made of carbon tape
and cloth costing up to $100 a pound.288 Steel is ~40¢ a pound. To be sure,
only ~15% of the cost of a finished steel car part is the steel, and composites can save much of the other ~85% through simpler shaping, assembly,
and finishing. As VW notes, low-cost carbon-composite automotive structures could cut the weight of a car by 40% (most firms say 50–65%289) and
body parts by 70%, making this approach “cost effective even if the
manufacturing costs per part are still expected to be higher”290—partly
because lighter cars need smaller, cheaper powertrains. But so vast was
the cost-per-pound gulf that automakers—who think of cost per pound,
not per car, and whose steel bodymaking skills are exquisitely evolved—
long denied carbon any serious attention. Until the mid-1990s, Detroit
had only a few dozen people exploring advanced composites, probably
none with manufacturing experience. The industry’s few high-profile
experiments, like Ford’s 1979 1,200-pound-lighter LTD sedan and GM’s
1991 Ultralite (Fig. 18a below), were interpreted mainly as proving carbon
cost far too much to compete, so it wasn’t worth learning more about.
Such skepticism is now waning.291 In 2002, Ward’s Automotive Weekly
reported that BMW “is planning to do what virtually every other major
auto maker on the planet has dreamed about: mass producing significant
numbers of carbon-fiber-intensive vehicles that are not only light and fast,
but also fuel efficient.”292 BMW “…in the last two years…has shown several concept vehicles using considerable amounts of carbon fiber, and two
projects are now in development for serial production. A major introduction with volumes as high as 100 cars per day is expected in 2005,” based
on the Z22 concept car revealed in 2000 (Fig. 13c) and using 200+ pounds
of carbon fiber. (BMW’s M3 CSL’s five-layer, 13-lb carbon-fiber roof is
already automatically made 5× faster than hand layup.293) BMW has built
“the world’s first highly automated production process for carbon-fiber-
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
reinforced plastic” and with 60 full-time composites manufacturing
specialists, is “going ‘all out’ to develop the skills…to expand use of the
material from its primary domain in motorsports…to regular vehicle production.”294 BMW itself says the Z22’s “tour de force,” whose “success
borders on the phenomenal,” has “virtually launched a technological
chain reaction at BMW…[which is] developing this material for use in
series production cars” because it’s “50% lighter than steel” and “performs extremely well in vehicle crash testing.”295
This transition is accelerating as firms like Hypercar, Inc.296 commercialize
low-cost advanced-composite manufacturing solutions. Its illustrative
patented process297 automatically and rapidly lays carbon and/or other
fibers in desired positions and orientations on a flat “tailored blank,”
compacting the layers, then thermopressing the blank into its final threedimensional shape. Long discontinuous carbon fibers can flow into complex shapes and deep draws with high strength and uniformity. The goal
is 80% of the performance of hand-layup aerospace composites at 20% of
their cost—enough to beat aluminum for an autobody with the same
attributes, and, in a whole-system solution, possibly to match steel per car
(net of savings from fewer pounds, smaller powertrain and other parts,
and simpler, more agile, and less capital-intensive manufacturing).
By unlocking advanced composites’ full manufacturing potential for
ultralight autobodies, such methods’ far lower capital, assembly, and
parts count could decisively favor early adopters.
Ultralight but ultrastrong Even using traditional manufacturing
processes, carbon fiber is showing up in 2004 hoods, roofs, and other
parts from U.S., Japanese, and German automakers, including Chevrolet,
Dodge, Ford, Mazda, Nissan, and BMW. Its virtues include ultralight
weight, virtual immunity to fatigue and corrosion, and impressive crash
absorption (Fig. 14).298 Advanced composites’ encouraging crash performance299 is made possible by three attributes:
• Optimally shaped carbon-fiber composite structures can absorb an
order of magnitude more crash energy per pound than steel or
aluminum (Fig. 15).
Saving Oil
BMW calls its
carbon-fiber Z22
concept car a “tour
de force” whose
“success borders on
the phenomenal.”
294. Including the world’s
largest Resin Transfer
Molding press (1,800
tonnes): Ponticel 2003.
295. BMW Group 2001/2002.
296. Hypercar, Inc.
(www.hypercar.com).
Please see the declaration
of interest on pp. 61, 63.
297. Cramer & Taggert 2002,
Whitfield 2004.
Ultrastrong carbonfiber composite
autobodies can save
oil and lives at the
same time, and by
greatly simplifying
manufacturing, can
give automakers a
decisive competitive
advantage.
298. iafrica.com 2003.
299. In principle, the
Mercedes SLR McLaren,
at a test weight of 1,768 kg
curb weight + 136 kg driver
and luggage, could dissipate its crash energy
against a fixed barrier at
29.3 m/s, 105 km/h, or 66 mph if its 6.8 kg of composite crush structures, reacting against a rigid member, absorbed a nominal 120 kJ/kg—half what the best
carbon-thermoplastic structures can absorb (Fig. 15). However, data from Larry Evans of GM, presented by Ross & Wenzel (2001, p. 18), suggest that over
99% of U.S. vehicle crashes, and 75% of fatalities, involve a velocity change of less than 35 mph.
• Such structures can crush relatively smoothly, not jerkily as metal does,
because rather than accordion-folding, composite crush structures can
crumble into dust from one end to the other (Fig. 16), using the crush
length or stroke ~1.5–2× as efficiently as metal can.300
300. Norr & Imbsweiler (2001) give a simple empirical example of a ~50% gain in crush efficiency.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
57
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
• A light but extraordinarily stiff beam can surround the passengers
and prevent intrusion.301
Bigger needn’t mean
heavier, lighter needn’t
mean smaller, and
light but strong materials can improve
both safety and fuel
economy without
tradeoff.
A car that’s bigger
but not heavier is safer
for people in both
vehicles.
Adding size without
weight provides
protection without
hostility.
These safety capabilities of advanced composites, combined with other
materials like aluminum and plastic foams, solve the mystery that led a
majority of the 2001 NRC panel to
Figure 14: The strength of ultralight carbon-fiber
doubt lightweight cars’ crashworautobodies was illustrated in November 2003
thiness and hence to limit analyzed
in Capetown, South Africa, when a Mercedes
SLR McLaren was rammed by a VW Golf
mass reductions to 5%.302
running a red light. The SLR—a 1,768-kg, handThey assumed that autobodies
layup, 626-hp, 207-mph, 16-mpg, street-licensed
will continue to be welded from
supercar priced at a half-million dollars—
sustained only minor damage despite being hit
stamped steel in today’s designs,
on the driver’s-side door (the photograph shows
so future cars, like past ones, will
a carbon side panel popped off).
become less protective if they’re
The unfortunate steel Golf , roughly one-fourth
lighter than the SLR , had to be towed.
made lighter and—importantly—
303
if nothing else changes. But as
Henry Ford said, “I cannot imagine
where the delusion that weight
means strength came from.”304
Cars’ weight does not necessarily
determine their size or crashworthiness as historic correlations
imply.305 Of course, physics requires
that, other things being equal, a heavier car is safer to be in but more dangerous to be struck by.306 But other things aren’t equal. Vehicle size—“the
most important safety parameter that doesn’t inherently conflict with
greater fuel efficiency”307—doesn’t shift risk from the projectile to the target
vehicle. Rather, a car that’s bigger but not heavier is safer for people in
301. Research recently assembled by senior University of Michigan physics professor Marc Ross indicates a shift in many safety experts’ opinions, especially in Europe, away from momentum and towards intrusion into the passenger compartment as the major cause of death and serious injuries. He points out
that although European designs that greatly improve safety, yet weigh very little, are not automatically applicable to the U.S.—where many victims weren’t
wearing seat belts and automakers must assume they might not be—some obvious design improvements could greatly reduce the risk and consequences of
intrusion. These include very strong and intrusion-resistant passenger cells, front ends homogeneous over a large enough area to spread loads (rather than
penetrating at a point, as light trucks’ framerails tend to do), and stiffnesses calibrated so that vehicles absorb their own kinetic energy rather than crushing
a lighter collision partner (though this requirement becomes more difficult if masses within the fleet vary widely). None of these is a current or contemplated
U.S. requirement; instead, NHTSA rules now encourage stiff front ends (to resist fixed-barrier frontal collisions) and may soon disincentivize downweighting.
If U.S. safety policy continues to diverge from findings and the policies that emerge from them elsewhere, U.S. cars may become increasingly unmarketable
abroad.
302. Members Dr. David Greene and Maryanne Keller correctly dissented: Greene & Keller 2002.
303. Kahane 2003.
304. Ford 1923, p. 14.
305. The October 2003 NHTSA report (Kahane 2003) was prepared in support of a 2004 rulemaking in which the Administration proposes to reclassify CAFE standards by weight classes, so that heavier cars need not be as efficient as lighter ones. This would tend to perpetuate and exacerbate the “mass arms race,”
increase public hazard, and make American cars less marketable abroad. The Technical Annex’s Ch. 5 further discusses this issue and its proper resolution.
306. Evans & Frick (1993), state two “laws” fitting their data for vehicles of various types and sizes but similar materials and design: that “when other factors
are equal, (1) the lighter the vehicle, the less risk to other road users, and (2) the heavier the vehicle, the less risk to its occupants.”
307. O’Neill 1995. The quotation refers specifically to crush length, which O’Neill agrees can improve safety in both striking and struck vehicle if added without increasing weight, i.e., by using light-but-strong materials. (His article also correctly criticizes a ten-year-old journalistic misstatement of some safety
principles of ultralight cars.)
58
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
SMC
natur-fibers
aluminum
steel
50
foam
100
honeycomb
150
C-FRP
200
thermoplastics (PEI)
250
The practical engineering reality of
safe lightweighting, coordinated
0
0 0 0 0 30 35 0 50
with sophisticated occupant protecfiber volume (vol. %)
tion, is well established,312 and is
Source: Herrman, Mohrdeck, & Bjekovic 2002, p. 17.
advanced enough, particularly in
313
Europe, that a 2.5-meter-long
Smart microcar’s crash dummies “survived” a 32-mph 50%-offset frontal
crash with a Mercedes S-class.314 Similarly, open-wheel race car drivers are
seldom killed by 200-plus-mph crashes in their ultralight carbon-fiber vehicles. Kinetic energy rises as the square of speed—it’s 13 times as great at 220
mph as at 60—so just the race drivers’ special padding can’t save them.
But Swedish driver Kenny Brack survived a horrific 220-mph crash in his
carbon-fiber race car (Fig. 17).
308. O’Neill, Haddon, & Joksch 1974. However, there are some special circumstances, such as a collision with a
deformable object, where weight may protect occupants. This aim too can be achieved by suitable design without simply
relying on high weight that increases hazard to struck vehicles.
309. Herrmann, Mohrdieck, & Bjekovic 2002. In practice, carbon-fiber layers in crush structures are often overlain or interwoven with tougher fibers (aramid, glass, Spectra®, polyethylene, etc.) for fracture control in extreme failures.
C-FRP (thermoplastics)
C-FRP (thermosets)
Figure 15: Advanced composites’ remarkable crash energy absorption
Carbon-fiber reinforced polymer (C-FRP) crush cones and similar structures
can absorb ~120 kJ/kg if made with a thermoset resin like epoxy,
or ~250 with a thermoplastic, vs. ~20 for steel.309 Crush properties can also be
optimized by mixing carbon with other fibers.
specific energy absorption
(progressive crushing) (kJ/kg)
both vehicles, because extra crush
length absorbs crash energy without adding the aggressivity of
weight. Since weight is hostile but
size is protective,308 adding size
without weight provides protection
without hostility. Lighter but
stronger materials can thus decouple size from both weight and safety: bigger needn’t mean heavier,
lighter needn’t mean smaller. Light
but strong materials can improve
both safety and fuel economy without tradeoff, offsetting ultralight
vehicles’ mass disadvantage.
50 50
Figure 16:
Two of these 7.5-lb
(3.8-kg) Mercedes
SLR McLaren crush
cones, whose crosssection varies over
their two-foot length
to provide constant
deceleration, can
absorb all of that
supercar’s energy
in a ~65-mph fixedbarrier crash,310
with 4–5× steel’s
energy absorption
per pound.311
310. Miel 2003.
311. Mercedes-Benz 2004a.
312. For example, after quoting a 91% weight saving for a carbon-composite square-section axial tube compared with an
equally energy-absorbing steel one, the Academy’s Commission on Engineering and Technical Systems (CETS 1999, p. 67)
concluded: “Based on these test results, the committee believes that structural concepts and analytical tools are available
to design lightweight PNGV concept cars that will perform safely in collisions with heavier cars because of the excellent
energy absorption characteristics of the alternate materials.”
313. E.g., Frei et al. 1997; Frei et al. 1999; Muser et al. 1996; Käser et al. 1995; Moore & Lovins 1995.
See also note 301, and the extensive efforts and publications of the EU’s Composit network, www.compositn.net.
314. Auto-Motor-Sport Journal 1999; Müller 2000; Three Point Motors Ltd. 2003.
An offset-crash video is at www.off-road.com/mbenz/videos/Sclass_Smart.avi, with a >2:1 mass ratio corresponding to a velocity far over 70 km/h;
an offset-crash still photo with an E-class is at Wolfgang 1998; and a head-on is at Pistonheads.com 1998.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
59
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
If safety required
weight,
bicycle helmets
would be made of
steel.
Two days after Brack’s crash, nearFigure 17: On 12 October 2003 at the Texas
ly all major media misreported in
Motor Speedway’s Chevy 500, 37-year-old race
car driver Kenny Brack’s car was hit from
unrelated stories that a federal stabehind, pinwheeled into the air and into a steeltistical study had found (as the
and-concrete barrier, and smashed to bits.
He survived this 220-mph crash with five fracNew York Times put it) that “reductures, because his ultralight carbon-fiber
ing vehicle weights would have a
Formula One racecar is extraordinarily strong
deadly effect over all.” It had
and designed to fail in a controlled fashion. He
recovered in six months, and by April 2004 was
found nothing of the sort: since it
expecting to return to racing later in 2004.
examined only old steel cars, it
He commented, “Obviously, this shows that
couldn’t predict anything about
the cars are very safe.”
future cars using different materials
and designs.317 If safety required
weight, Brack would be dead and
bicycle helmets would be made of
steel. Brack is alive because carbon
fiber is such a great crash-absorber.
If you doubt the strength of ultralight structures, just try, without tools,
eating an Atlantic lobster in its shell. Carbon composites are even stronger.
315
316
315. Cavin 2004.
316. Savage 2003.
317. Not only was the
NHTSA study cited (Kahane
2003) artfully framed to
invite readers to draw the
erroneous conclusion that
the media did—that its
backward-looking correlations showed what was
possible with future cars—
but those correlations were
deeply flawed: e.g., see
Public Citizen 2003; Ross &
Wenzel 2001; and Wenzel &
Ross 2003. The last of these
shows that Kahane’s
weight/risk correlation may
even be artifactual, since a
safety/quality correlation,
measured by resale value
or quality ratings, explains
the data better.
318. Gladwell 2004, which
notes, for example, that the
subcompact VW Jetta has
half the total fatality rate
(per million car-miles driven) of popular SUVs nearly
twice its weight, and protects its occupants similarly
or better. See also Roberts
2001; Bradsher 2002.
319. Ross & Wenzel 2001;
Wenzel & Ross 2003. These
authors correctly emphasize that weight should be
reduced not only via light
materials but also via
improved body design and
styles and via high-efficiency powertrains.
Impressively safe ultralight composite family vehicles have already been
designed. Some show promise of cost-competitiveness. National policy
should encourage such decoupling of size from weight and safety, reverse
the spiraling (and unsubtly marketed) “arms race” in vehicle weight,318
and save both lives and oil. Nothing in the backward-looking federal
study, nor in science or engineering, refutes such innovations. To make its
light-vehicle fleet safer, the U.S. needs to
…resolve the incompatibility of light trucks with cars [in weight, frontal stiffness, and height] and it needs to continue development and adoption of powerful crash mitigation and avoidance technologies. Making heavy vehicles lighter
(but not smaller) and making lighter cars larger (but not heavier) would not only
increase safety but also increase fuel economy…[by] over 50%….319
Or as GM’s former head researcher on crash safety confirmed in
March 2004:320
Increasing the amount of light-weight materials in a vehicle can lead to lighter,
larger vehicles [possessing]…all of the following concurrent characteristics:
reduced risk to its occupant in two-vehicle crashes; reduced risk to occupants in
other vehicles into which it crashes; reduced risk to its occupants in single-vehicle crashes; reduced fuel consumption; reduced emissions of [CO2]….
Advanced polymer composites are especially attractive for this role,
because not only do they have exceptional crash energy absorption, stiffness, and durability, but they also hold promise of simpler and cheaper
manufacturing with dramatically reduced capital intensity and plant scale.
320. Evans 2004, his bullets
converted into semicolons.
60
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Applying ultralight materials As automakers start to exploit this
new ability to make cars ultralight but ultrasafe—and more agile in
avoiding crashes—they’re also making them superefficient and fun to
drive. Four carbon-fiber concept cars instructively combine all these
attributes (Fig. 18). GM’s pioneering 1991 Ultralite packed the interior
space of a Chevy Corsica (twice its weight and half its efficiency), and
the wheelbase of a Lexus LS-400, into a Mazda Miata package, matching
the acceleration of that era’s 12-cylinder BMW 750iL, yet with an engine
smaller than a Honda Civic’s. In 2002, Opel’s diesel Eco-Speedster turned
heads with 155 mph, 94 mpg, and below-Euro-4 emissions.321 Toyota’s
2004 Alessandro Volta is a muscular carbon-fiber hybrid supercar with
compact-car fuel economy.322 And less widely noted but of particular technical interest is the production-costed and manufacturable virtual design
for the Revolution concept car (Box 7)—a midsize crossover-style upmarket SUV. SUVs’ share of U.S. light-vehicle sales rose from 1.8% in 1975 to
26.1% in 2004,323 so this popular category merits especially close analysis.
Saving Oil
Concept cars,
early market cars,
and components
presage an era of
affordable carbonfiber cars that
combine better safety
with sporty performance and startling
fuel economy.
Figure 18: Four carbon-fiber concept cars: From L–R, they are: 1991 GM 4-seat Ultralite (635 kg, Cd 0.192, 0–60 mph in 7.3 s, 84 mpg [2.8
L/100 km] using a nonhybrid gasoline engine); 2002 Opel 2-seat Eco-Speedster diesel hybrid (660 kg, Cd 0.20, max. 155 mph [250 km/h], 94
mpg [2.5 L/100 km], below-Euro-4 emissions; 2004 Toyota Alessandro Volta, 3 seats abreast, by-wire, 408-hp hybrid, 32 mpg, 0–60 mph in
<4 s, top speed governed to 155 mph; 2000 Hypercar Revolution show car that mocks up a midsize SUV virtual design (857 kg, 5 seats,
by-wire, Cd 0.26, 0–60 mph in 8.2 s, 114 mpg-equivalent with fuel cell).
The design of Revolution revealed important new information for this
study, but our discussion of it requires an explanation and a declaration
of interest. This concept car was designed in 2000 by Hypercar, Inc., a
small private technology development firm supporting the auto industry’s transition (p. 57). The senior author of this report (ABL) invented
the broad concept of such vehicles in 1991 and has written and consulted
about it extensively. He is cofounder, Chairman, and a modest stockand option-holder of Hypercar, Inc. and is cofounder and CEO of Rocky
Mountain Institute, this report’s nonprofit publisher, which became a
321. Automotive Intelligence News 2002; Car.kak.net 2003; Paris Motor Show 2002. The Euro 4 standard permits four times the particulates and ten times the
NOx allowed by the next stage of very stringent U.S. standards (Tier 2, bin 5), but technical solutions appear feasible (Schindler 2002). Starting in October
2003, Mercedes-Benz introduced additional particulate filtration, requiring no additives, to several of its European car lines. However, care must be taken in
interpreting these solutions’ effectiveness because the Euro 4 test does not catch or count many of the extremely fine particulates that U.S. regulators consider of greatest health concern—a concern that is increasing (Cavanaugh 2004). German automakers are seeking to accelerate Euro 5 standards.
322. See Toyota, undated (downloaded 12 April 2004); Serious Wheels, undated; RSportCars.com, undated. This Italdesign concept car was revealed at the
Geneva Motor Show in February 2004. The midengine hybrid system, with four traction motors, is variously said to be, or to be derived from, the Lexus RX
400h’s powertrain. The even faster 2004 Chrysler concept ME412 has been simulated to achieve 248 mph (0–60 mph in 2.9 s) with a carbon-fiber body and
850-hp engine (Composites World 2004).
323. ORNL 2003, Table 4.9.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
61
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
7: Superefficient but uncompromised
The original Revolution concept car (Fig. 18d)
was designed in 2000 by an engineering team
assembled by Hypercar, Inc., including two
leading European Tier One engineering companies (a noted UK design integrator with
automaking and Formula One sister firms, and
the German company FKA325 for powertrain) and
such industrial partners as Michelin and Sun
Microsystems.326 The goal was to demonstrate
the technical feasibility and the driver, societal,
and automaker benefits of holistic vehicle
design focused on efficiency and lightweight
composite structures. It was designed to have
breakthrough (5–6×) efficiency, meet U.S. and
European safety standards, and satisfy a rigorous and complete set of product requirements
for a sporty and spacious five-passenger SUV
crossover vehicle segment with technologies
that could be in volume production at competitive cost within five years. The team developed
a CAD (computer-aided design) model of the
concept vehicle and performed static bending
and torsion, frontal-crash, powertrain-performance, aerodynamic, thermal, and electrical
analyses to validate the conceptual design, as
well as midvolume manufacturing and cost
analyses described below (see end of this box
on p. 63, Fig. 20, and pp. 65ff).
The design combined packaging comparable to
the 2000 Lexus RX 300 (five adults in comfort,
and up to 69 ft 3 (1.95 m3) of cargo with the rear
seats folded flat), half-ton hauling capacity up a
44% grade, and brisk acceleration (0–60 mph in
8.2 s). Its low drag and halved weight yielded a
simulated EPA adjusted 114 mpg (2.06 L/100 km),
or ≥99 mpg on-the-road (2.38),327 using a fuel cell
~5 percentage points less efficient than today’s
norm.328 Industry-standard simulations also
showed that a 35-mph (56-km/h) crash into a wall
wouldn’t damage the passenger compartment—
62
most cars get totaled at about half that speed—
and that even in a head-on collision with a steel
SUV twice its weight, each at 30 mph (48 km/h),
the ultralight car would protect its occupants
from serious injury.329 The highly integrated
design process systematically maximized “mass
decompounding”—snowballing of weight savings—by downsizing and even eliminating components and subsystems that a capable and
comfortable but ultralight vehicle wouldn’t need.
Revolution ’s direct-hydrogen fuel-cell system
was simulated to achieve a 330-mile (531-km)
(continued on next page)
325. Fahrkraftwesengesellschaft Aachen mbH, www.fka.de.
326. Revolution development was led by D.F. Taggart, the same aerospace engineer who’d led the 1994–96 Lockheed Martin Skunk Works®
team that designed a 95%-carbon advanced-tactical-fighter airframe—
one-third lighter but two-thirds cheaper than its 72%-metal predecessor, due to its clean-sheet design for optimal composites manufacturing—mentioned on p. 82.
327. As is common industry practice, FKA simulated Revolution’s onroad efficiency in 2000 by multiplying each vector in the EPA driving
cycles by 1.3. This achieves an on-road result, ~95 mpg, conservatively
below the ~109 mpg obtained by successively applying the EPA’s normal
correction factor from “laboratory” to “adjusted” combined city/highway efficiency—a reduction in efficiency by factors of 0.9 for city and
0.78 for highway driving—and EPA’s ~0.96 estimated adjustment to realworld driving. Before these adjustments, using unscaled driving vectors,
the “laboratory” efficiency of this fuel-cell variant was 134 mpg on the
EPA combined city/highway cycle.
328. The original 2000 fuel-cell Revolution design’s peak-load efficiency
was reduced from 50% to 45% in the fuel-cell one of the three variants
simulated for this study. However, since peak loads seldom occur, this
reduced the average adjusted EPA mpg by only 1 mpg (while reducing
cost considerably), and required no mass adjustment. Acceleration was
increased 12.5% to EIA 2025 projected values by scaling the front traction
motors, buffer battery, and power electronics, not the fuel cell. The resulting fuel-cell system efficiency averaged 59.4% city and 58.7% highway, or
tank-to-wheels, 55.0% and 46.4%, respectively. In contrast, 50% peak-load
efficiency is assumed for a nominal neat-hydrogen fuel-cell system in the
latest MIT Energy Lab analysis (Weiss et al. 2003). All these efficiencies
are on the same accounting basis: from neat-hydrogen Lower Heating
Value to fuel-cell DC bus output, net of all auxiliary power consumption.
329. Defined as meeting the same deceleration limits as the Federal
Motor Vehicle Safety Standards require for a 30-mph fixed-barrier
frontal crash test. The project’s limited budget precluded simulation of
other crash modes, let alone physical crash tests, but the quality of the
simulation tools and the Tier One prime contractor’s extensive Formula
One experience give reasonable confidence that the design would perform well in crash modes other than the two basic ones simulated.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
major stockholder on spinning off Hypercar in 1999. (Sir Mark MoodyStuart, author of the Foreword on p. xviii, is also a stockholder.) Hypercar
has not done automotive design since 2000, and since 2002 has focused
solely on composite manufacturing processes. Normally such a history of
self-interest would discourage more than passing mention of Revolution’s
design. But in fact, that design’s technical details are uniquely useful here,
for four reasons: it integrates advanced technologies more fully than any
market vehicle; it shows promise of competitive manufacturing cost; its
technical description, unlike that of advanced automaker projects, has
been rather fully published; 324 and its proprietary cost and other data,
available to RMI only because of these historic connections, permit far
deeper analysis than could be performed on any automaker’s advanced
projects. For these reasons, Revolution has been specially reanalyzed for
this report, both by and independently of RMI, using both public and
proprietary data, to produce a detailed case-study of advanced technolo-
Box 7: Superefficient but uncompromised (continued)
average range on 7.5 lb (3.4 kg) of safely stored
compressed hydrogen in U.S.-approved 5,000psi (350-bar) carbon-fiber tanks; newer Germanapproved tanks tested by GM operate at twice
that pressure, which would extend the range
beyond 500 miles (by 2003, 10,000 psi had
become an industry design norm). Revolution
combined a body much stiffer than a good
sports sedan’s 330 with all-wheel fast digital traction control, 13–20-cm variable ride height, and
smart semiactive suspension, so it should be
very sporty. Revolution may have been the first
car designed from scratch to be all-digital, allnetworked, with all functionality in software—a
highly robust computer with wheels, not a car
with chips—thus potentially offering many new
330. Static analyses showed bending stiffness 14,470 N/mm, torsional
stiffness 38,490 N•m/deg, first bending mode 93 Hz, and first torsion
mode 62 Hz. These are respectively 85%, 221%, 140%, and 141% of corresponding values for the ultra-high-strength-steel ULSAB-AVC midsize
car design, whose 218-kg body-in-white is 17% heavier. The large-area
adhesive bonding in the Revolution’s body would also maintain stiffness
throughout the very long life of the vehicle, vs. most metal autobodies’
rapid degradation of spot-welds.
331. Probably therefore overstated; ULSAB-AVC, for example, was
assumed in its cost analysis to gain a 10% reduction in suppliers’ bids
via “virtual negotiation.”
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
A 35-mph crash
into a wall wouldn’t
damage the passenger compartment and
even in a head-on
collision with a steel
SUV twice its weight,
each at 30 mph, the
ultralight car would
protect its occupants
from serious injury.
324. Cramer & Taggart 2002;
Lovins & Cramer 2004;
www.hypercar.com.
kinds of aftermarket customization, wireless
background tune-ups and diagnostics, remote
upgrades, and other novel value propositions.
Since its body wouldn’t fatigue, rust, nor dent in
a 6-mph collision, and the rest of the car was
radically simplified, the design was consistent
with a 200,000-mile warranty.
In 2000, Hypercar, Inc. analyzed detailed costs
for hypothetical 2005 production of Revolution at
a greenfield plant making 50,000 vehicles per
year—about the volume of the aluminum Audi
A2 . A 499-line-item Bill of Materials, developed
in close collaboration with two Tier Ones and
the supply chain, accounted for 94% of manufacturing cost. It supported cost estimates
based 60% on unnegotiated331 quotations by the
supply chain in response to anonymous Tier
One cost-pack requests, 6% on standard partsbin costs, 25% on consultant analyses of powertrain cost, and 9% on in-house analyses of proprietary production processes for composite
components. The resulting production-cost private analysis then supported the extensions
summarized in Fig. 20 on p. 65.
63
Saving Oil
A specially optimized
ultralight design for
a superefficient
midsize SUV provides
an archetype for
saving 69% of 2025
light-vehicle fuel at a
cost of 57¢/gallon.
In case the composites
don’t work out, lighter
steel autobodies offer
a worthy backstop.
332. Strict economic marginalists might want us to
stick with a lightweight nonhybrid, saving 58% of its
gasoline at 15¢ per pretax
gallon, and say it’s uneconomic to go to a lightweight
hybrid, saving 72% of its
gasoline at an additional
cost of $2.36/gal. We bundle
these together and go
straight to the hybrid
because its savings, as an
integrated package, cost
only a very profitable
56¢/gal, and it’s impractical
to buy a non-hybrid car and
then retrofit it to a hybrid.
Buying only 81% of the
hybrid’s potential savings
would be a classic suboptimization. Energy-saving
experts call it “cream-skimming”—buying only a smaller, cheaper increment of
savings in a way that makes
larger savings, though costeffective when bought
together, economically and
practically unobtainable
when bought separately.
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
gies’ integrated performance and cost. Similar results could be readily
achieved by others, and all our calculations are shown in and replicable
from Technical Annex, Ch. 5. Using this particular foundation for our
analysis simply makes it more specific, detailed, and transparent than
could otherwise be possible. That one of us has been so long engaged
with it, and has advanced it in other fora, should not disqualify it from
informing this report.
Lighter-but-safer vehicles dramatically extend cheap oil savings
The important lesson of our light-vehicle analysis, summarized next, is
that advanced-composite ultralight hybrids roughly double the efficiency
of a normal-weight hybrid without materially raising its total manufacturing cost. That’s partly because of simpler and cheaper manufacturing,
partly because ultralighting shrinks powertrains.
The model-year-2005 hybrid SUVs shown in Fig. 5 (p. 31) use Prius-style
hybrid powertrains to double their efficiency, but they don’t yet capture
the further efficiency of cutting their weight in half. This synergistic
ultralight-hybrid combination, illustrated by the simulated Revolution
hybrid, is the approach we adopted for all State of the Art light vehicles.332
Figure 19: Two-thirds of a State of the Art light vehicle’s fuel saving comes from its light weight
This figure charts causes of successive reductions in nominal 2025 gal/y of gasoline consumed, comparing a Revolution hybrid to a 2004 Audi Allroad 2.7T base vehicle. About 68% of our State of the Art hybrid’s
fuel saving comes from its 51% lighter curb weight.333 The hybrid powertrain—the main focus of the better
of the previous studies—contributes only ~16% of the savings, and so do the reductions in drag, rolling
resistance, accessory loads, etc. that previous studies partly or mostly counted.
The exact shares will depend on the sequence assumed for these savings, but lightweighting will always
dominate, and should be done (along with other tractive-load reductions) before hybridization in order to
make the powertrain smaller, simpler, and cheaper. The key is ultralight weight; without it, two-thirds of
the potential fuel savings are lost.
461
956
105
111
279
base vehicle
curb weight –51%
aero, tires, etc.
hybridization
light hybrid
Source: RMI analysis (see Technical Annex, Ch. 5).
333. For this graph, we estimate the mass effect using the method and data of An & Santini (2004) and scale it to the Audi 2.7T base vehicle; obtain the drag/rolling-resistance/accessories/integration term from the difference between the hybrid and internal-combustion-engine Revolution variants; and use the Revolution hybrid to obtain
the last decrement of fuel use. The calculation is in Technical Annex, Ch. 5.
64
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Lightweight steels could also achieve much of the same oil saving more
conventionally if advanced composites prove unready (p. 67).
Two-thirds of what
makes ultralight
hybrids so efficient
and cost-effective is
their light-but-strong
construction.
Our analysis highlights
an important new area
of the design space—
ultralight,
superefficient,
exceptionally safe,
and cost-competitive.
Two-thirds of what makes ultralight hybrids so efficient and cost-effective
is their light-but-strong construction, as shown in Fig. 19. Previous studies
largely neglected this major opportunity, for three reasons: presumed
high cost from using light metals, presumed loss of crashworthiness from
not offsetting lighter weight with disproportionately strong and highenergy-absorption new materials, and lack of an integrated design that
properly captures ultralight autobodies’ snowballing weight savings and
powertrain downsizing. (Previous studies also generally assumed partby-part substitution, which largely misses these key benefits and often
raises new technical complications.334) By avoiding all three problems,
our methodology highlights an important new area of the design space—
ultralight, superefficient, exceptionally safe, and cost-competitive.
Specifically analyzing and integrating this approach’s breakthrough
potential for each type and size of light vehicle is so daunting that
334. Brylawski & Lovins 1995.
Figure 20: An ultralight hybrid SUV saves 72% of today’s comparable model’s fuel at 56¢/gal. Comparison of pretax retail price and
other attributes of a 2004 Audi Allroad 2.7T with Tiptronic SUV and three functionally comparable Revolution simulated SUV/crossover
vehicles—gasoline-fueled with internal-combustion-engine (ICE) and with hybrid drive, and fuel-cell-powered. Details of this RMI
analysis, based on efficiencies simulated by independent consultants, are in Technical Annex , Ch. 5. Note that the composite autobody
(dark blue) is only a small part of Revolution’s total vehicle cost, and the major efficiency gains of ultralighting are quite inexpensive.
Three all-wheel-drive Revolution variants with
EIA 2025 acceleration (0–60 mph in 7.1 s)
gasoline
ICE
gasoline
hybrid
hydrogen
fuel cell
Revolution
Audi
Allroad 2.7T
Pretax retail price of selected crossover vehicles (2000 $)
curb mass (kg)
1,929
878
916
892
gasoline hybrid
EPA adjusted mpg
18.5
44.6
66.0
107.8
hydrogen fuel cell
Cost of
Saved Energy*
(2000 $/gal)
relative to:
Audi Allroad 2.7T
previous step
0.
–
0.15
0.15
0.56
2.36
2.11
12.32
pretax price
over Audi (%)
1.6
7.4
31.9
mpg increase
over Audi (%)
140
256
481
decrease in
annual fuel use
@13,874 mi/y
(2025 SUV) (%)
58
72
83
Revolution
2004 Audi Allroad
2.7T w/Tiptronic
gasoline ICE
0
10000
20000
30000
40000
50000
60000
Audi dealer cost
(component costs
and profit unavailable)
factory final assembly labor
body and structure
markup to dealer invoice price
interior trim and
instrument panel
markup to MSRP
factory overhead and rent
destination charge
chassis and HVAC
electrical and electronics
six dealer options (to make
comparable to Revolution)
propulsion
* 5%/y real discount rate, 14-y life, 0.10/y implicit capital
recovery factor
Source: RMI analysis based on proprietary production costs (Box 7, p. 63).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
65
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Advanced-composite
ultralight hybrids
roughly double
the efficiency of a
normal-weight hybrid
without materially
raising its total
manufacturing cost.
nobody has attempted it. One would need to create and simulate a complete virtual design separately optimized for each of about a dozen main
subclasses. This would cost tens of millions of dollars, so the results
would doubtless be secret. Yet we have been able to approximate such
results adequately and transparently by adapting Hypercar, Inc.’s
published and proprietary data on its virtual design for the Revolution
ultralight midsize SUV fuel-cell crossover vehicle (p. 61 and Box 7).
We engaged Hypercar and independent consultants to analyze that
design’s efficiency and cost using gasoline-engine powertrains instead
(Box 8), while meeting EIA’s 2025 assumed new and stock vehicle efficiency, vehicle-miles, acceleration, and other attributes. This yielded the results
in Fig. 20. We then extrapolated those results to all 2025 light vehicles,
by the methods in Boxes 9–10 (Technical Annex. Ch. 5), to obtain the fleet
results shown in Fig. 21.
Based on these data for lightweight hybrids, we adopt as State of the Art
an average light-vehicle fuel saving of 69% at $0.57/gal. Fig. 21 compares
these ultralight vehicles to the conventional, small, incremental improvements shown earlier in Fig. 11. It also shows the 2000 Revolution design’s
Figure 21: 1990–2004 comparison of absolute mpg vs. incremental costs for new U.S. light vehicles: ultralighting doubles the savings.
The studies (curves) and the two market vehicles (red) shown in Fig. 11, contrasted with the Revolution crossover-vehicle virtual design
(dark green), this report’s findings (light green, 2025 sales mix), and the steel industry’s virtual design (magenta). Prior studies didn’t
consider the potential for ultralight designs to save more fuel at lower cost. Note that the Revolution hybrid, and the State of the Art
light vehicles inferred from it, all cost about the same as today’s ordinary-weight Prius hybrid. This means that advanced composites’
fuel- and life-saving advantages—opening up the new design space on the right-hand half of the graph—are roughly free.
The ULSAB-AVC 52-mpg gasoline-internal-combustion-engine steel design (magenta) illustrates another path to saving fuel, and implies
the scope for a more efficient hybrid version whose attributes RMI roughly estimates as shown. Please see text for citations.
All vehicles shown in green are adjusted to EIA’s 2025 acceleration capability for that class of vehicle
(treating Revolution as a small SUV). RMI's 2004 average vehicles are for EIA’s 2025 sales mix.
NRC High
2001 light
trucks
4,500
4,000
NRC Low
2001 light
trucks
3,500
3,000
2,500
2,000
2004 Prius
(2004 actual
to ~2007 goal)
DeCicco & Ross 1995 Full Avg
NRC
Low
2001
cars
2004 RMI State of the Art
average
light truck
2000 Revolution w/AWD
hybrid powertrain
1,500
2002 ULSAB-AVC hybrid
(rough RMI estimate of
initial and more mature cost)
2004
RMI
State of the Art
average car
DeCicco, An, & Ross 2001 Mod & Adv cars
2004 RMI Conventional Wisdom average car
1992 VX subcompact
90
80
2002 ULSAB-AVC
50
40
30
2000 Revolution
w/AWD ICE
70
500
60
1,000
0
NRC
High
2001
cars
2004 RMI
Conventional
Wisdom
light truck
20
price increase (MSRP 2000 $)
5,000
absolute miles per U.S. gallon
(EPA adjusted, combined city/highway)
Source: RMI analysis described in text, pp. 62–73 and Technical Annex, Ch. 5.
66
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Saving Oil
Revolution ’s passenger cell’s 14 self-fixturing composite parts
underlying variants from which
are easily hand-liftable and snap together for gluing.
these fleet potentials are derived
This and its colorable-in-the-mold body shell make the body shop disappear
(all adjusted in 2004 to 2025 EIA
and the paint shop optional—the two costliest elements of automaking.
acceleration), as well as the light(See Box 10, p. 73.)
steel ULSAB-AVC virtual design.
This comparison reveals the dramatic advantages of lightweighting
hybrids—using either advanced composites, as we assumed,
or new lightweight steels, which are heavier but cheaper.
Our Conventional Wisdom vehicles are also shown, comfortably within the
zone of the conservative NRC curves. But since they save less fuel than a
non-hybrid gasoline-engine Revolution, and cost the same or more to
build, they’re not as economic a choice as State of the Art vehicles.
335–340. See Box 8.
Some readers may not be comfortable with our calculations of cost based
on the Revolution analysis (Technical Annex, Ch. 5), or may view differently
the technological and economic risk of switching automaking to advanced composites. For those who expect automaking to remain in the
Iron Age of which it is the highest achievement, the global steel industry’s
ULSAB-AVC design with Porsche Engineering (p. 55) offers a conventional-materials alternative for doubling fuel economy and improving crashworthiness at no extra cost. This conventional technological backstop for
our less familiar composites route should greatly reduce the perceived
technical and economic risks of achieving breakthrough efficiency at reasonable cost. Specifically, based on the 52-mpg gasoline-engine Taurusclass ULSAB-AVC design,360 which is 21% heavier than the 45-mpg gasoline-engine crossover Revolution but 19% lighter than a 2004 Prius, one
could reasonably estimate that a mature hybrid 2WD ULSAB-AVC would
sell for a few thousand dollars more than the steel industry’s projected
~$20,260 (2000 $) for its non-hybrid versions, and would get ~74 mpg,361
approaching the same supply curve of savings vs. cost as the similar-size
State of the Art average car. This confirms that ULSAB-AVC lightweight
steel hybrids should be considered plausible backup candidates and
worthy competitors for major oil savings in case advanced composites
proved unable to fulfill their promise.
CAUTION: ENTERING CALCULATIONAL THICKET.
The next six pages summarize the analysis whose results we just presented in Fig. 21.
Box 8 shows how we analyzed the efficiency of the gasoline version of the Revolution ultralight SUV.
Box 9 explains how we extended its cost analysis (Fig. 20, p. 65) to all light vehicles.
Box 10 discusses light-vehicle cost comparisons more broadly.
If you don’t need this level of detail—far more than this report presents for any other use of oil,
because light vehicles account for 46% of 2025 oil use and three-fifths of potential oil savings—
you are now finished with light vehicles and can skip to the next biggest oil use, heavy trucks, on p. 73.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
341–353. See Box 9.
354–359. See Box 10.
360. We assume a curb
mass of 1,059 kg (Porsche
Engineering 2001, Ch. 10, p.
20) rather than the apparently dry-and-bare mass of
998 kg stated elsewhere.
Prius is 1,311 kg, the nonhybrid AWD Revolution 878
kg. We assume that the
costs and prices in Shaw &
Roth 2002 are in 2001 $.
361. For mpg, we scale 52
mpg by the 1.44 Prius/
ULSAB-AVC ratio of tonmpg, and subtract 1 mpg
for mass compounding. We
estimate a range of costs in
two ways. First, we scale
the $1,832 marginal cost of
electrically hybridizing the
AWD gasoline Revolution
(adjusted downward somewhat for 2WD, upward
slightly for mass compounding) by the 1.15
ULSAB-AVC /Revolution ICE
test-mass ratio to get perhaps $2,000. Second, we
scale the 1Q2004 (2000 $)
~$3,740 Prius marginal
MSRP of hybridization (D.
Greene, personal communication, 25 March 2004)
downward by the test-mass
ratio, yielding ~$3,160 but
expected to fall by half in
the next few years (id.).
More elaborate calculations are of course possible
but probably not useful
without a full design
exercise.
67
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
We assumed our midsize SUV would achieve the same muscular 0–60 mph in 7.1 seconds
that EIA assumes for light trucks in 2025.
8: Analyzing an ultralight hybrid’s efficiency
The 2000 virtual design described in Box 7 (pp.
62–62) used a fuel cell, but this study requires a
detailed understanding of how such an ultralight, low-drag vehicle would work with gasoline
powertrains. Rocky Mountain Institute therefore
commissioned independent reanalyses of what
fuel economy this Revolution platform could
achieve using a gasoline engine (Technical
Annex, Ch. 5).335 RMI’s consultants first boosted
acceleration by 12.5%, to 0–60 mph in 7.1 s, to
match EIA’s light-truck forecast for 2025,
although comparable 2003 U.S. crossover vehicles already accelerate 20% faster than cars
and 22% faster than light trucks. This made the
114-mpg 2000 fuel-cell Revolution 4% heavier
and 4% less efficient (109 mpg). The consultants
then simulated an all-wheel-drive powertrain
using Honda’s ~2000 Insight 1-L Otto engine,336
because it’s light and efficient, has a published
performance map, and was designed for a car
with the same curb weight and drag coefficient.
(The 2000 5-seat carbon-fiber Revolution fuelcell SUV weighs the same—857 kg (1,888 lb)—
as the 2-seat aluminum Insight.) The resulting
gasoline-fueled Revolution variants were simulated337 to get 62.4 mpg with a hybrid powertrain,
or 45 mpg with a non-hybrid automated 5-speed
transmission.338 Making the hybrid Revolution ’s
powertrain as efficient as a 2004 Prius ’s would
increase its fuel economy to 68.4 mpg,339 which
we then reduced to 66 mpg to adjust for the
assumed all-wheel drive.340
335. The consultant on fuel economy was FKA (note 325), probably the leading Tier One engineering firm for advanced powertrain design and simulation.
Mass and cost were analyzed by Whole Systems Design, Inc. (Boulder, CO), whose principal, car engineer Timothy C. Moore, had previously led RMI’s
and Hypercar’s automotive simulations and was a key contributor to Hypercar’s cost analyses during the Revolution project. He previously designed
four cars (two by himself), built three, sold one, and won national awards for three.
336. Scaled linearly in torque. Conservatively, no adjustment was made in the efficiency map to reflect the somewhat higher efficiency that a larger
engine would normally achieve.
337. Using a second-by-second proprietary model (Longitudinal Simulation model of ika/fka Rev. 2003) frequently engaged by major automaker clients of
FKA, and correcting for hybrid battery state-of-charge and for the powertrain variants’ weight, engine size, and other characteristics. See Technical
Annex, Ch. 5, for details.
338. Neither powertrain was fully optimized, so the hybrid Revolution’s tank-to-wheels efficiency averaged 30.4%—like a 2003 Prius’s 31%—rather than
a 2004 Prius’s 33.2%. Specifically the laboratory (unadjusted dynamometer) 2004 Prius measurements are: city (synthesized from 43% cold and 57% hot
testing), 66.6 mpg, 35.0% engine efficiency, 36.2% system efficiency (tank-to-wheels); highway, 64.8 mpg, 34.6% engine efficiency, 30.1% system efficiency. System efficiency exceeds engine efficiency (An & Santini 2004) because of regenerative braking, which is 68.1% efficient on the city and 64.1% on
the highway cycle. On the adjusted EPA combined city-highway cycle, which reduces laboratory city mpg by 10% and laboratory highway mpg by 22%,
then combines their reciprocals, 2004 Prius thus gets 55.3 mpg at a system efficiency of 33.2%. These Prius data were generously provided by D. Hermance (Executive Engineer, Toyota USA), personal communications, 14 and 20 April 2004.
339. The reverse calculation from physics yields the same answer, as follows. If the Revolution hybrid design had the mass, aerodynamic drag, rolling
resistance, and accessory test load of the 2004 Prius, it would get an adjusted EPA mpg of 50.7 mpg, vs. Prius’s actual 9.2%-better 55.3 mpg—reasonably close to the 9.3% greater tank-to-wheels efficiency of the tested Prius vs. the simulated Revolution. Thus a Revolution hybrid with a Prius powertrain yields the same mpg as a Prius with Revolution physics and accessory test load, as shown in Technical Annex, Ch. 5. (It’s inappropriate to resize
Revolution’s powertrain for increased tractive load with Prius physics because the Prius powertrain already has a higher specific power than the
Revolution hybrid’s conservative mass budget assumed: e.g., Prius’s battery, at 1.25 kW/kg, weighs 58% less per kW than Revolution’s ~1995-vintage 0.53
kW/kg, which would save the 2004 Revolution hybrid 27 kg.) Prius’s better powertrain efficiency includes both hardware and software effects, but since
the incremental pretax retail price calculated below for a Revolution hybrid’s powertrain is comparable to or below Prius’s actual marginal powertrain
price (see note 178), any incremental powertrain hardware cost of the more efficient Prius over the Revolution hybrid can be considered negligible.
340. This 2.4-mpg reduction, from the senior author’s and a consultant’s engineering judgment, is believed to be very conservative because the traction
is electric, the powertrain highly regenerative, and the AWD mass already included. Unlike the larger loss with mechanical drivetrains (e.g., the 2004
Audi A6 3.0 loses 2.0 mpg or nearly 9% adjusted EPA efficiency between the 2WD and AWD versions), the Revolution’s AWD conversion incurs only the
minor electric, inertial, and mass penalties of using four traction motors rather than two.
68
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Saving Oil
9: Analyzing and extending ultralight vehicle costs
To extend to other light vehicles the efficiency
and cost directly simulated for our Revolution
archetypical State of the Art ultralight hybrid,
we combined previous work with special new
analyses, and made simplifying assumptions
that adequately capture the first-order effects
to be characterized. Our main guide in this
extension was Ashuckian et al.’s detailed 2003
efficiency and cost analysis of potential gains in
light-vehicle efficiency in California, applying a
consistent and well-documented framework to
many different reports and vehicle classes.341
Details are in Technical Annex, Ch. 5.
Ashuckian et al.’s analysis contains a detailed
study by one of the most respected industry
analysts, K.G. Duleep (Energy & Environmental
Analysis, Inc. [EEA]), that forms the basis of our
Conventional Wisdom light vehicles. (For that
calculation we start with EIA’s projected mpg by
vehicle class in 2025; scale each mpg figure by
341. Ashuckian et al. 2003.
342. EIA’s NEMS outputs, kindly provided to us by EIA’s modeling
experts, show numbers of vehicles, adjusted EPA fuel economy, estimated on-road fuel economy, and miles traveled. Unfortunately, vehicle-miles traveled and vehicle populations are shown only as aggregated totals, not by vehicle class or subclass. We therefore had to
construct a vehicle stock-and-flow model, shown and tested in
Technical Annex, Ch. 5, to yield this level of detail needed to support
our analysis.
343. This consultant appropriately adjusted and resized components,
and conservatively applied replacement drivesystem component costs
from three proprietary industry sources. These are mass estimates
developed for Hypercar, Inc. by Lotus Engineering (Norwich, UK); a
Bill of Materials for actual components in Lotus’s Elise production
sports car, which has power and mass broadly comparable to
Revolution’s; and a confidential teardown of an existing production
vehicle with certain features and capacities also similar to
Revolution’s. The iterative process of adjusting mass and propulsion
power requirements went through three recursions for the hybrid and
two for the other two variants, adjusting each time such elements as
engine, cooling systems, electric motors, traction batteries, transmission, differentials, fuel storage and delivery systems, and exhaust systems. Secondary elements were appropriately adjusted by informal
estimate. The fuel-cell variant’s stack was downsized (per kW) from
the 2000 design, saving cost, reducing peak-load efficiency by five
percentage points (note 328, p. 62), and leaving mass unchanged.
Details are in the Technical Annex , Ch. 5.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
the ratio of EEA’s policy case to EEA’s base case
for that vehicle class; and adopt the incremental capital costs calculated by EEA for its policy
case, converted to 2000 $. The result, if fully
applied to EIA’s forecasted light-vehicle fuel use
in 2025, would be a 27% fuel saving at $0.53/gallon for cars and $0.34/gallon for light trucks.)
Our State of the Art calculations are necessarily
less direct. Any procedure short of designing
from scratch an optimized vehicle of each class
is necessarily imperfect, but we believe our
methodology, described next and qualified in
Box 10, realistically captures the key relationships between vehicle classes as well as the
recent progress in hybrid and lightweight vehicle technologies uniquely embodied in the
Revolution design. The results are summarized
in Fig. 20. On a consistent accounting basis, the
hybrid Revolution’s 3.5× (47-mpg) efficiency gain
over Allroad 2.7T yields a 72% fuel saving at an
incremental capital cost of $3,190. The Cost of
Saved Energy is $0.56/gallon at our 5%/y real
discount rate and EIA’s 2025 vehicle-mi/y.
To obtain this result, we first constructed a
base-case model of the light-vehicle fleet from
2000 through 2025 by subclass, matching EIA’s
forecasted fuel use within a few percent, to
support analyses not otherwise possible using
EIA’s limited model outputs.342 Then we commissioned Whole Systems Design (WSD, Boulder,
CO) to update and expand Hypercar, Inc.’s proprietary 2000 cost analysis.343 From a 13-vehicle
universe of near-peer vehicles (Technical
Annex, Ch. 5), WSD and Hypercar, Inc. experts
chose Audi’s 2004 Allroad 2.7T with Tiptronic as
today’s market vehicle closest to Revolution in
function, performance, features, and price344—
(continued on next page)
69
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Unlike previous studies, which often sought to extrapolate from midmarket cars to all other vehicle classes,
our analysis applies to the fleet an integrated heart-of-the-new-market SUV-crossover design
specifically tuned to the most demanding requirements, including safety, size, comfort, and performance.
9: Analyzing and extending ultralight vehicle
costs (continued)
hence a comparable baseline from which we
could estimate incremental costs and efficiencies. Drawing on a variety of studies validated
across a wide range of vehicle classes and
used to convert manufacturing cost to dealer
invoice price and to MSRP, we applied the highvolume Ford Explorer’s 79% markup from manufacturing cost to invoice price and the low-volume Audi Allroad 2.7T ’s 10.4% markup from
invoice price to MSRP. We included destination
charge (the weight-related factory-to-dealer
shipping cost) but not sales tax. Of the hybrid
Revolution’s calculated incremental pretax retail
344. In 2000 $, the Audi lists at $34,256 MSRP plus $5,102 of options
for feature comparability, 6 hp/100 lb, 0–60 mph in 7.6 s, and 18.7
mpg. The comparison is inexact but appears to understate
Revolution’s relative value.
345. To calculate costs and energy savings, we compared the hybrid
Revolution’s 66-mpg fuel economy to the projected fuel economy
for EIA’s “compact SUV” vehicle class in 2025 (21.9 mpg). We treated
the increase in fuel economy between the 2004 Audi Allroad and
EIA’s 2025 compact SUV baseline—equivalent to 20% of the total fuel
intensity reduction from the Audi to the Revolution hybrid—as a
business-as-usual efficiency improvement that should happen in the
absence of changes in government policy and business practice,
which is EIA’s standard and legally mandated assumption. We therefore assumed that 20% of the $3,190 incremental cost for the hybrid
Revolution relative to the Audi is attributable to efficiency improvements that would have happened anyway, and subtracted that percentage of the cost (totaling $646) to calculate net incremental costs
for the compact SUV category of $2,544. This approximation
assumes a more or less linear relationship between fuel intensity
and cost for the limited improvements considered, which also implicitly include any evolution in feature set over those 21 years.
346. We used that study’s “full hybrid” case because it reflects
significant weight reduction, though not nearly as much as Revolution ’s, and is most comparable to our State of the Art case even
though its hybrid powertrains are two generations beyond that of the
2004 Prius to which we adjusted the hybrid Revolution’s efficiency.
347. This is a standard analytic assumption used by NAS/NRC (2001)
among others. According to ORNL (2003), the mean car in 2001 was
9 and the mean light truck 8 years old, both growing; for the most
recent model year available (MY1990), median survival was 17 years
for cars and 15-odd for light trucks, also both growing as reliability
and corrosion resistance continue to rise.
70
price of $3,190, we assumed 20% (in line with
EIA’s assumptions) would reflect the next 21
years of spontaneous business-as-usual
improvements, yielding a net incremental cost
of $2,544.345
Scaling these results to other vehicle classes
shows that the Revolution hybrid’s technology
and cost translate into saving 69% of the 2025
light-vehicle fleet’s fuel at a Cost of Saved
Energy of $0.53/gallon for cars and $0.59/gallon
for light trucks. To obtain that result, we
assumed that the fuel economy and incremental
cost relationships between vehicle classes in
the ACEEE full-hybrid case (summarized in
Ashuckian et al.) would also apply to our State
of the Art case. The ACEEE case contains both
hybrid powertrains and significant lightweighting, making it a reasonable proxy. We assumed
that our hybrid Revolution efficiency of 66 mpg
and incremental cost of $3,190 was comparable
to ACEEE’s “compact SUV” category, then
scaled mpg and incremental costs to the other
vehicle classes proportionately.346 We applied
the resulting savings in each vehicle class
to EIA’s projected 2025 vehicle-miles, using a
14-year new-vehicle life.347
Our cost analysis conservatively omitted three
further opportunities:
• It made no allowance for efficiency gains that
may come from making the Revolution design
more similar to mainstream vehicles in acceleration, accessories, and other features.
In these and other respects, Revolution is truly
a luxury design, but we lacked the details
and resources to analyze potential efficiency
gains from a more “stripped-down” version.
(continued on next page)
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
Box 9: Analyzing and extending ultralight vehicle costs (continued)
• Any Otto-engine vehicle can gain another
~25–30% in mpg-gasoline-equivalent
(~40–45% in mpg of diesel fuel) by switching
to a modern diesel engine if, as European
automakers believe, such engines can be
made to meet emerging U.S. emission rules.
Surprisingly clean and quiet diesels are now
entering the U.S. market, e.g. in the 2005
Mercedes E320 CDI, which meets 45 states’
current emission standards, is hoped to meet
the rest by 2006,348 and is 24% more efficient.349
Diesel-hybrid concept cars are being
designed not just for muscle (Figs. 18b, c) but
also for fuel economy, like Toyota’s ultralight
2001 ES 3 subcompact’s 77 mpg gasolineequivalent.350 U.S. automakers would find
diesel adoption convenient, and it’s just passing 50% market share in Europe, but we
entirely omitted this major option until air
issues are definitively resolved.
• To convert EPA laboratory mpg to actual
on-road mpg, this study applied EIA’s “degradation factor” (which rises to 0.801 for cars
and 0.792 for light trucks in 2025) to reflect
EIA’s projected congestion, driving patterns,
highway speeds, and other real-world shifts.
That yields a 2025 estimated on-road mpg
about four percentage points below EPA
adjusted mpg. This four-point loss may be
appropriate for the inefficient vehicles EIA
assumes, but significantly understates the onroad efficiency of the far more efficient vehicles we’ve analyzed.351
Together, we believe these conservatisms, plus
the considerations in Box 10, more than offset
the inevitable uncertainties in our analysis. We
are also comfortable with our use of an SUV
crossover design—combining the attributes of a
rugged off-road vehicle and a luxury sports
sedan—as a surrogate for a fleet increasingly
dominated by those attributes. In actual use,
most SUVs are now car substitutes—perhaps
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
1–13%352 go off-road or do heavy towing—and
so are ~75% of pickup trucks.353 Unlike previous
studies, which often sought to extrapolate from
midmarket cars to all other vehicle classes,
our analysis applies to the fleet an integrated
heart-of-the-new-market midsize SUV-crossover design specifically tuned to the most
demanding requirements, including safety.
348. According to Mercedes executives paraphrased by W. Brown
(2004). See also Foss 2004 and Marcus 2004. The tighter emission rules
that now prevent this model from being sold in California (and NY, MA,
VT, and ME, which use its standards) will apply nationwide from 2007.
349. I.e., 24% excluding or 40% including the 13% greater energy of a
gallon of diesel fuel than of modern gasoline.
350. This 733-kg plastic-and-aluminum 4-seat subcompact has low
drag (Cd 0.23) and frontal area (1.98 m 2). Its ultracapacitor-storage
hybrid drive peps up its 1.4-L turbocharged diesel engine and boosts
EC-cycle efficiency to 88 mpg of diesel fuel, equivalent to 78 mpg of
gasoline: Toyota 2001; EV World 2001.
351. This is for three reasons: our State of the Art vehicles’ lower aerodynamic drag improves their fuel economy at EIA’s increased future
highway speeds; their hybrid drive recovers braking energy in EIA’s
increasingly congested urban driving, typically making them more efficient in the urban than the highway mode (rather than about one-third
less as EIA assumes); and their accessory loads are far lower (e.g., at
least fivefold for the Revolution design)—a major source of the difference between adjusted-EPA and on-road mpg, since EPA’s test procedure leaves air-conditioning and other accessories turned off. The derivation of EIA’s AEO04 degradation factors is described by DACV 2000.
Unfortunately EIA’s 2025 fleet is not technically characterized—just its
driving patterns—so we can’t calculate the adjustment needed. Adding
a further level of confusion, EIA’s “EPA rated” mpg figures in the NEMS
database turn out (J. Maples, EPA, personal communication, 10 May
2004) to mean NHTSA CAFE mpg, which are similar but not identical to
EPA laboratory mpg; we use the latter to avoid the NHTSA data’s varying <1-mpg distortions by model vs. calendar years and by adjustments
for alternative-fuel-capable vehicles. EIA’s “EPA rated” mpg are not, as
one might assume, EPA adjusted mpg. EIA’s degradation factor of ~0.8
in 2025 therefore combines the conventionally separate steps of converting from EPA laboratory to EPA adjusted mpg (by subtracting 10%
from city and 22% from highway laboratory values) and then correcting
further to obtain estimated on-road mpg. To fit this unusual and unstated EIA convention, our underlying analysis in Technical Annex, Ch. 5
starts with EPA laboratory mpg, then translates the results into the EPA
adjusted mpg used there and in this report.
352. Bradsher (2002, pp. 112–114), citing industry interviews and surveys. However, nearly half the intending buyers of SUVs and pickups
surveyed in 1998 said they planned to drive off-road: Steiner 2003,
Table 5.1.7. (Some, perhaps many, may have thought that meant driving
on graded dirt or gravel roads.) There appear to be similar, perhaps
smaller, discrepancies between intended or actual purchase of towing
packages and their use. Such gaps would not be surprising, since
SUVs are among the most heavily marketed U.S. products, and that
marketing emphasizes rugged off-road fantasies.
353. DeCicco, An, & Ross 2001, p. 16.
71
Saving Oil
Option 1. Efficient use of oil: Transportation: Light vehicles: Advanced automotive technologies
10: Comparing light-vehicle prices
It’s hard to compare rigorously the costs and
prices of light vehicles even in today’s market,
let alone in 2025’s. However, combining accepted, validated, and documented analytic methods
can yield a reasonable and probably conservative estimate of the pretax retail price of
advanced-composite ultralight hybrid-electric
vehicles.
Feature differences prevent exact comparisons
between current market vehicles and
Revolution-like concept designs. Revolution’s
standard equipment is equivalent to options354
that add $5,390 (2004 $) to Allroad ’s MSRP. Some
features are simply different: instead of leather
seats (a $3,600 extra on Allroad), Revolution has
mesh seats like the market-leading Herman
Miller Aeron chair’s, and genuine (not faux) carbon-fiber trim. Revolution isn’t designed for towing, which relatively few users need (often the
same ~5% who actually drive off-road), while
Allroad can tow 3,300 pounds. But Revolution
provides adjustable ride height and exceptional
stability control. On balance, Revolution’s feature package can fairly be interpreted as adding
thousands of dollars to today’s SUV base-model
MSRPs. It’s therefore conservative to assume
rough feature parity between the advanced
2000 Revolution we costed and a well-loaded
2004 luxury sport-utility crossover vehicle. But
several further broad comments help put such
comparisons in perspective.
The minor price differences, in either direction,
of radically more efficient but uncompromised
light vehicles are within the range of normal
trimline variations in today’s market.355 For example, the 2000 Ford Explorer’s MSRP ranged from
$20,495 for a two-door, 2WD base version to
$34,900 for a fully loaded four-door, 4WD Limited
Edition.356 This $14,405 spread exceeds any plau-
72
sible price premium for even the costliest (fuelcell) State of the Art efficiency gain, pessimistically assessed. Yet the inherent feature-richness
of advanced vehicles, the customer attributes
inherent in hybrid powertrains (smoothness,
quiet, low-end torque, abundant onboard
power capacity, etc.), and the potential to offer
“exciters”357 should allow automakers to create
appealing customer option packages, analogous
to but surpassing those of the 2004 Prius, that
simply fold in superefficiency as an incidental
coproduct of market-leading vehicle innovation.
RMI’s analysis of the cost of saving fuel, as if
that were all that such advanced vehicles offer,
invites the narrowest sort of economic comparison between ways of achieving that single goal.
In fact, if that were customers’ sole criterion for
buying cars, only one model would survive in
each vehicle class. The global auto industry
actually sells a dizzying and dynamic array of
models and options because people are complicated and have many kinds of preferences.
From a marketing perspective, the public reaction to Revolution convinces us that most
buyers will want such vehicles for a great many
reasons, differing only in where fuel economy
comes into their list of priorities, so it’s not
important how they rank mpg among desired
attributes.
(continued on next page)
354. These are the $3,600 Premium package, $750 cold weather package, $1,100 premium audio package, $850 telematics package, $390
tire-pressure monitoring system, $1,350 navigation system, and $950
17” wheels.
355. This insight is due to DeCicco, An, & Ross (2001, p. 20).
356. DeCicco, An, & Ross 2001, p. 20.
357. This is car marketers’ term for exciting features buyers didn’t
expect a car could provide, such as the Japanese Prius model’s ability
to parallel-park itself. A car that perfectly implements everything
expected of it, but no more, will generally undersell a model that
adequately implements all expected features but also implements
an “exciter” even in half-baked fashion.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation
Heavy trucks
Class 8 trucks—18-wheelers and their kin, with a Gross Vehicle Weight
Rating (GVWR) of 33,001 to 80,000 lb—were the fastest-growing user of
highway transportation fuel during 1991–2001 (3.6%/y), outpacing even
light trucks (3.4%/y).362 Our analysis shows a surprisingly inexpensive
potential to raise Class 8 trucks from EIA’s 6.2 to 11.8 363 miles per gallon
of diesel fuel (State of the Art), and then, with four further truck-specific
improvements plus system-benefits, to ~16 diesel-mpg equivalent.
The trucking industry is intensely competitive and worries constantly
about fuel (its second-biggest factor cost, whose price correlates closely
362. ORNL 2003, Table 2.6.
363. Although we adopt EIA’s 2000 baseline of 6.19 miles per gallon of diesel fuel, new Class 8 trucks in and after MY2004
may now be getting ~5.9 (implied by Kenworth 2003, p. 18) or even fewer (by some anecdotal accounts) mpg on-road.
(continued on next page)
Box 10: Comparing light-vehicle prices (continued)
New U.S. cars’ real prices have risen steadily for
30 years amid business-cycle fluctuations. If the
linear-regression trend continued, average real
car prices (2000 $) would rise by 11.5% or $2,290
during 2000–2010, and by $5,725 during 2000–
2025 358 (three times EIA’s assumed $1,765 or 9.6%
rise for average light vehicles during 2000–2025).
Historic price growth, driven by consumer preferences and incomes and by regulatory requirements, has so far swamped any price growth
due to efficiency technologies, and would continue to do so if extrapolated.359 But the steady
growth in expectations, capabilities, and features
that customers (and often regulators) expect
also tends to be delivered by technologies in or
closely related to the efficiency portfolio. Price
increases due to saving fuel without also adding
other desired attributes are likely to be small.
These cost and price comparisons omit
Revolution-like vehicles’ biggest economic implication: the way they transform automakers’
risk/reward ratio by dramatically cutting capital
investment, assembly effort and space, and parts
count. Revolution’s passenger cell’s 14 self-fixturing composite parts are easily hand-liftable and
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Systematic improvements in weight, drag,
engines, tires, and
other technical details
can double heavy
trucks’ efficiency at
an average cost under
25 cents a gallon,
and may save twothirds with smarter
regulatory policies.
Omitting these technical improvements
costs a trucker
four times as much as
installing them.
snap together for gluing. This and its colorablein-the-mold body shell make the body shop disappear and the paint shop optional—the two
costliest elements of automaking. Product cycle
time can also be slashed, especially with further
progress toward soft tooling that could ultimately
replace the half-billion-dollar football-field-full of
up to a thousand progressive steel die-sets that
take a thousand engineers two years to design
and create. (Molding the Revolution’s composite
body takes only 14 die-sets—fast and cheap to
make—and they run at low pressure to mold
polymers, rather than stamp steel.) Revolutionlike designs could apparently compete on cost
and earn market returns, but their strategic agility, capital-leanness, and much smaller minimum
production scale may be far more important,
offering early adopters a new path to striking
competitive advantage.
Finally, we would reemphasize the inexorable
progress of technology, which on past form
(Fig. 11) seems likely by 2025 to surpass even
our hopes, and to surpass by far our 2004
technologies.
358. DeCicco, An, & Ross 2001 p. 21 and Fig. 4.
359. DeCicco, An, & Ross 2001, p. 21.
73
Saving Oil
Option 1. Efficient use of oil: Transportation: Heavy trucks
Michigan’s biggest
food-grade tanker
fleet went from 5 to
12 mpg-equivalent
without our proposed
technological
improvements.
with bankruptcies). While truckers care intensely about their rigs’ look
and feel, trucking lacks some of the emotional and marketing complications of changing car design, and tractor makers tend to have shorter
product development cycles and faster adoption of innovations than do
automakers. Yet trucking has specific structural barriers (pp. 150–154
below) that have constrained adoption of many important fuel-saving
technologies, creating major opportunities if these barriers are systematically overcome. Our technological analysis (Technical Annex, Ch. 6)—firmly grounded in industry experiments and in detailed studies at MIT and
two National Labs—finds that full adoption could save ~38% of EIA’s
forecasted 2025 truck fuel, at a cost below $1 per saved gallon of diesel
fuel at the nozzle, or 41% at below $2/gal. These savings rise respectively
to 46% and 50% when we add four further truck or driver improvements,364 whose hard-to-determine cost we conservatively assume to equal
the average of the previous gains.
Taken together, the technical potential could save diesel fuel at an average cost of 25¢/gal fuel, vs. EIA’s projected 2025 partly-taxed price 365 of
$1.34/gal. Average costs per saved barrel of crude oil range from –$1.2 to
+$3.7, and the costliest (though tiny) increment of savings costs $26.3–
32.9/bbl.366 The technological savings total ~1.0 Mbbl/d of crude oil in
2025, nearly 4% of EIA’s total projected demand. Although only ~20–25%
of the technological improvements are retrofittable, a new-truck buyer
would pay only about one-fourth as much to install them as not to.
(note 363 continued from
previous page) That’s
because emission-reducing
technologies hastily added
by seven makers of heavytruck diesel engines
reduced their fuel efficiency. These technologies
avoided even costlier EPA
This analysis also doesn’t assume seven regulatory changes367 (whose savpenalties (several thousand
dollars per truck) for noning or cost or both we found hard to estimate accurately) that could subcompliance with 2002 nonstantially raise the 2025 saving to perhaps 55–65% of EIA’s 2025 heavymethane hydrocarbon and
NOx emissions standards
truck fuel use. The most important regulatory change—safely raising
set in 1997, under a 1999
GVWR to the 110,000 lb allowed in Europe (Canada allows 138,000 lb)—
consent decree settling
government claims that
would raise load per trip by ~53% and cut fuel per ton-mile by ~15–30%,
they had installed illegal
cut emissions and congestion, and enable international shipments to be
software to turn off engine
emission control systems
fully loaded at origin.368 When Michigan adopted a 164,500-lb limit for its
during highway driving
(EPA, undated[a]).
Unfortunately, the resulting fuel-cost penalty to truck owners was greater, especially high fuel prices. The same happened to MY2004+ urban buses.
By not counting this recent development, and in effect assuming the engine-makers will restore their engines’ pre-2004 efficiency while meeting the emissions standards, our analyzed efficiency improvements may be understated.
364. These were (1) allowing a 1-axle increase, (2) navigation technology to reduce wasted miles, (3) real-time fuel-economy display, and (4) better driver’s
education to optimize acceleration, deceleration, and match of operating conditions to powertrain map (note 376).
365. Including federal and state but not county or local taxes, and assuming 15 ppm sulfur.
366. The cost per saved barrel is negative if the cost of saving the retail diesel fuel is less than the cost of converting crude oil at the refinery entrance into
diesel fuel at the retail pump, because then the refinery would have to use cheaper-than-free crude oil in order for the value chain to deliver diesel fuel more
cheaply than saving it.
367. These seven omitted regulatory options are: (1) increase maximum GVWR to the European norm of 110,000 lb, (2) further utilize the truck-rail-truck modal
shift by stacking-train rail and rail-to-truck transloading, (3) federally increase trailer lengths from 53 to 59 feet, (4) federally increase trailer height to 14 feet
from 13.5 feet—already part of some states’ standards, (5) allow double and triple trailer combinations, accompanied by disc brakes giving more brakes-perpound-of-GVWR, (6) reduce speed limits to 60 mph, and (7) reduce empty miles by consolidating loads with large carriers. We estimate that these savings
would cumulatively account for at least another 20–40% off any given baseline truck-stock fuel intensity.
74
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Heavy trucks
Figure 22: Supply curves for saving oil by improving on EIA’s 2025 heavy trucks
In 2003, on-road diesel prices (2000 $) averaged $0.89/gal pretax and $1.43/gal taxed; in May 2004, taxed prices exceeded $1.63.
Cost of Saved Energy
(2000 $/gal diesel)
2.50
Average
State of
the Art
CSE =
$0.25/gal
diesel
2.00
EIA 2025 post-tax diesel price ($1.34/gal)
1.50
EIA 2025 pretax diesel price ($1.04/gal)
Average
Conventional
Wisdom CSE =
$0.13/gal diesel
1.00
aerodynamics
engine
fuel cell auxiliary power
tires
transmission
0.50
weight reduction
0.00
(0.50)
0.00
0.25
0.50
0.75
1.00
1.25
diesel fuel saved (Mbbl/d)
from EIA Reference Case in 2025 with full adoption
Source: RMI analysis, EIA 2004.
roads, its biggest food-grade tanker fleet increased load per daily trip by
2.5×, equivalent to raising efficiency from 5 to 12 mpg, without any of the
technical improvements analyzed here. Safety and roads needn’t suffer,
because brake effectiveness and the number of axles would increase at
least in step with the vehicle’s total weight.369
Heavier loads
needn’t harm roads
or safety.
Figure 23: The evolution of heavy-truck tractors
L–R: typical ~5-mpg (diesel fuel) Peterbilt 379, ~7.5-mpg370 Kenworth T2000, PACCAR center-console concept tractor,371
and engineer/artist’s impression (commissioned by RMI) of a State of the Art lightweight, highly aerodynamic tractor.372
368. Major global shippers, such as a major automaker, have complained that the largest volume of material they ship is air—the empty space in partlyloaded containers.
369. Safety can be sustained and indeed increased, due mainly to better brakes (more stopping power per unit weight) and reduced tire-blowout risk (from
better wide-based single tires and automatic pressure control). Road damage depends not on GWVR but on the pressure resulting from distributing that
weight through axles to tires and road surfaces, so higher GWVR with even more axles can mean lower axle weight and road damage, though in some circumstances it could require bridge upgrades (Luskin & Walton 2001). It may be attractive to offer higher GWVR as part of a package limiting speed to, say,
60 mph (improving safety and saving additional fuel). Exterior airbags might be added too. Some engineers also favor “bullet truck” designs coupling tractor
and trailer in ways that could even eliminate the risk of jackknifing. (As illustrated at www.e-z.net/~ts/ts/jack.htm, a rig jackknifes when when the tractor’s
drive (rear) axle locks its brakes, making the tractor and trailer unsteerable and typically causing a rollover or making the rig sweep across multiple lanes of
traffic—the main cause of multi-vehicle pileups involving heavy trucks.)
370. PACCAR 2001; PACCAR 2002.
371. This concept design enhances both fuel efficiency and safety. The wedge shape of the cab reduces drag while providing front underrun protection.
The driver’s seat is located on the centerline of the vehicle for enhanced visibility. Photo copyright 2004, courtesy PACCAR Inc.
372. Original RMI artwork commissioned from Timothy C. Moore, Whole Systems Technologies, Inc. (Boulder CO); see note 335.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
75
Saving Oil
Option 1. Efficient use of oil: Transportation: Heavy trucks
At 65 mph,
a heavy truck’s
tractive load
is about
3 / 5 air drag,
2 / 5 rolling
resistance.
Fig. 22 shows the attractively low cost of the main technologies (Technical
Annex, Ch. 6). Most are analogous to those for light vehicles373 and are on
the market or under test by top firms. In order of decreasing fuel savings:
• Increase from 5 to 6 axles (as in Canada and Europe), improving
trucking (and much container-ship and rail-freight) productivity by
perhaps 15–20%—and incidentally easing the security-inspection burden by reducing the number of international container shipments.
373. What’s different is
mainly the allocation of
• Fully adopt the incremental engine improvements already planned,
vehicle losses. Lavender,
to achieve a brake thermal efficiency of 55% (DOE’s 2012 goal, vs.
Eberle, and Murray (2003)
today’s 40–44%), via variable-pitch turbos, turbocompounding, disprovide Woodrooffe &
Associates data showing
placement on demand, variable valve timing/lift, common-rail, piezothat a nominal Class 8 base
injectors, 42-V electrical systems, conversion of hydraulics to electrics,
truck driving on a level road
at 65 mph and a respectbetter lubricants, camless diesels, hybrid drive, and homogeneous
able 6.6 mpg uses 400 kW
charge compression ignition.
of fuel, of which 60% goes
to engine losses, 3.8% to
• Comprehensively apply the past 30 years’ aerodynamic improvements
accessory loads, 2.3% to
powertrain losses, and the
to reduce drag. Cd was ~1.0 in the 1970s, reaches ~0.6–0.7 today
remaining 34%, or 136 kW,
(Fig.
22b), and in advanced designs (Fig. 23d) can probably approach
to tractive loads comprising
the ~0.25 of today’s best market cars.
85 kWh aerodynamic losses
and 51 kWh rolling resistance (i.e., 5/8 aero, 3/8
• Use lighter, stronger, more durable tractor and trailer materials to save
tires). Since this analysis
5% of fuel—both by saving fuel through lower rolling resistance even
assumes constant speed,
braking losses are neglectwhen hauling light, bulky goods, and by carrying heavier payloads
ed; for an over-the-road
per trip (saving trips) when hauling heavy, dense goods.374
truck these are small but
not zero. Hill-climbing ener• Make air-conditioning and other auxiliary loads efficient, often
gy must also be counted,
and isn’t recovered except
electrically- rather than shaft-driven, and powered by an efficient APU
on roller-coaster roads or
(auxiliary power unit).
with hybrid drive. Ignoring
these refinements for actu• Fully apply superefficient wide-based single tires and automatic tireal driving cycles, the analysis suggests that saving
pressure controls.375
one kWh of tractive load
saves 2.5 kWh of fuel—
• Use loadsensing cruise control, reduced out-of-route miles, real-time
less than the ~7–8-fold
fuel-economy and gear-optimizing display for driver feedback, and
wheels-to-tank multiplier
for standard light vehicles
better driver education.376
(p. 52) because of the
big truck diesel engine’s
nominal 40% efficiency.
The authors (two from
National Labs) suggest an efficiency goal of 10.3 mpg.
374. Kenworth Trucking Co. (2003) provides data implying that in “some applications such as specialized truckload carriers and tank truck carriers,” weight
savings can be worth over 30 times as much for their increased payload as for their reduced rolling resistance.
375. Properly designed and operated, this combination needn’t increase blowout risk. The automatic inflation system not only saves fuel, but also provides
instant low-pressure warning, usually giving the driver time to pull over. Although there’s only one tire per axle per side, it should run cooler due to
decreased rolling resistance (two rather than four sidewalls are flexing). In any event, the apparent redundancy of a double tire is often illusory because
failure of one tire in a pair immediately overloads and blows out the other. Moreover, dual tires can develop unequal pressure and diameter because the
inner tire is closer to inboard brakes and can experience less air circulation. Most truckers who use wide-based single tires prefer them for smoother handling, easier maintenance, and better weight distribution as well as significant savings in weight (a half-ton per rig), fuel, and per-wheel taxation (Kelley 2002;
Davis 2002; Kilcarr 2002).
376. Uchitelle (2004), for example, describes CR England Inc.’s classes for its 3,800 professional drivers, who had often run their 2,600 heavy trucks’ diesels at
1,500 rpm rather than at the more fuel-thrifty 1,200 rpm.
76
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Heavy trucks
Saving Oil
Figure 24: The shape of the future? European concept trucks in tanker and (R) linehaul tractor variants
by Prof. Luigi Colani.377
Fig. 24 illustrates such aerodynamic and lightweighting gains in a
2001–02 concept tanker truck reportedly 43% more efficient than Europe’s
6.4-mpg class average—and, at 11.2 mpg, 90% of the way to State of the
Art’s 12.5-mpg target without using all its techniques, notably for engines.
Medium trucks
Class 3–7 trucks (GWVR 10,001–33,000 lb) use ~7% of highway-vehicle
fuel, and comprise a wide range of non-household trucks and vans.
They can generally use the same technologies already analyzed for light
and heavy trucks, because nearly all of them closely resemble the trucks
in one or the other of these categories.378 Technical Annex, Ch. 7, shows that
full use of those technologies in 2025, netted against an assumed 5–6%
penalty in fuel economy for emission controls, would save 45%
(Conventional Wisdom) to 66% (State of the Art) of Class 3–7 trucks’ gross
fuel use at, respectively, $1.23 and $0.85 per gallon of gasoline or diesel
fuel. The savings are so large partly because the fleet has so far improved
so little in mass, drag, or powertrains, and partly because its typical
urban-dominated driving cycle yields especially large savings from
hybrid drive. For example, Fedex expects the hybrid drives now being
deployed into its 30,000 OptiFleet E700 light-medium trucks to save 33%
of fuel and 90% of emissions—and this fleet is still using conventional
platform physics. DOE’s Advanced Heavy Hybrid Propulsion System
Program targets 50% fuel savings. Our analysis relies mainly on MIT and
Argonne National Laboratory studies, and assumes by 2025 the same
halving of hybrid powertrains’ extra cost that Toyota and Honda expect
for their hybrid cars over the next few years.
Non-household
trucks and vans
can apply light- and
heavy-vehicle
technologies to save
two-thirds of their
fuel at $1 a gallon.
377. Professor Luigi Colani,
whose design studio is
well known in the Germanspeaking world, is a
Sorbonne-trained aerodynamicist whose biography
lists consultancies for many
automakers and aerospace
firms. The tanker shown at
Spitzer Silo-Fahrzeugwerke
GmbH, undated and aluNET
International, undated, a project of Spitzer Silo-Fahrzeugwerke GmbH, is one of a family of seven developed since 1977. An over-the-road version is said to
have been tested at 20.9 L/100 km, and, at FleetWatch 2002, is said to represent a 43% saving from the European fleet average for its class. AkzoNobel 2002
reports the test (at the Bosch track in Boxberg) to have shown a ~30% fuel saving, but it doesn’t say compared to what; Professor Colani (personal communication, 12 May 2004) says it was a standard Mercedes truck, but further details are unavailable. Additional tractor and tractor-trailer images are at Colani,
undated; www.colani.de, undated; Bekkoame Co. Ltd. 2002; Autotomorrow 1989; www.lkw-infos.de; Luigi Colani Design, undated; and Virtualtourist.com 2003.
Schröder (2002) mentions the tractor in one test as 3.96 m high, 2.52 m wide, and 6.16 m long, with a 12.9-L 430-hp diesel engine. An article about the bodywork (www.handwerk-ist-hightech.de, undated) says Professor Colani’s AERO 3000 was modified from a DAF 95 XF. He informed the senior author [ABL]
orally of plans to build by autumn 2004, with a Dutch partner, a superefficient heavy truck which he said he expects to achieve Cd ~0.2.
378. Light trucks—Class 1 (≤6,000 lb) and Class 2 (6,001–10,000 lb)—accounted for 71% of total Class 1–8 truck fuel usage in 1997, and Class 8 for 22%
(ORNL 2003, Table 5.4). Class 6 (19,501–26,000 lb) used 4% and Class 7 (26,001–33,000 lb), which behave much like Class 8 heavy trucks, used 1.4%.
Class 3–5 trucks use only 22% of Class 3–7 truck fuel, which in turn is only 7% of highway-vehicle fuel, so they don’t need detailed analysis here.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
77
Saving Oil
Option 1. Efficient use of oil: Transportation
Fully adopting
proven innovations
in road management
could save 1–2%
of highway fuel
and a great deal of
highway-building
money and drivers’
time.
Intelligent highway systems (IHS)
Since 2001 congestion cost ~$70 billion
per year, and EIA
traffic forecasts
would make it worse,
intelligent highway
systems probably
save enormously
more money than they
cost, but we took
no account for that
wealth creation.
379. Intelligent
Transportation Society of
America, www.itsa.org.
380. The deferral value of
road investments and
avoided accidents would
modestly increase the
value of saved time. We
also chose to count IHS
effects outside the supplycurve context so as to
avoid the methodological
awkwardness of having to
put it near the left side of
the curve (because of its
negative cost), but thereby
leaving readers to wonder
how far it changes the
light-vehicle analysis.
We preferred to show the
light-vehicle analysis by
itself, unperturbed by
reduced driving.
78
The U.S. National Intelligent Transportation Systems Architecture contains 32 technical and operational options in eight categories. They save
congestion, driver time, fuel, money, and often pollution and accidents.
We analyzed some of the most obvious opportunities for saving fuel by
these means. Full adoption of incident management, signal coordination,
ramp meters, and electronic toll plazas on all major roads in 75 U.S. metropolitan areas’ would have saved 0.95 billion gallons of the 5.7 billion
gallons of fuel wasted by congestion in 2001. Adding a modest amount of
advanced routing technologies, and a small part of the potential offered
by a diverse additional technical portfolio described in Technical Annex,
Ch. 8, would have saved another 0.5 billion gal/y in 2001. This 1.45-billion-gallon saving potential in 2001 matches ITS America’s379 2002–2012
goal, scales to 1.68 billion gal/y (0.9% of total oil consumption) for all
highway vehicles in 2025, and is our Conventional Wisdom case.
We also considered six other major technologies that could save between
17-plus and 45-plus percent of the fuel otherwise wasted by congestion.
These include signal priority modeling for bus rapid transit and trucks,
intelligent cruise control, very close vehicle spacing, vehicle classifiers,
routing algorithms, and agent-based computing infrastructure, all
described and referenced in the final section of Technical Annex, Ch. 8.
Some of these measures are relatively costly, so we assume only the
cheapest one-fourth of the whole portfolio, whose composition will vary
with local circumstances. When deployed along with the Conventional
Wisdom technology suite, those least-cost additional technologies round
out our State of the Art IHS portfolio. Estimating its fuel savings and costs
is particularly difficult, so rather than simply adding up the savings from
every option, we conservatively estimate the impact of any subset to be
twice the total of the Conventional Wisdom portfolio; the actual savings
could be far greater.
The costs of both IHS portfolios assumed here are unknown but probably
modest, and are at least an order of magnitude smaller than the societal
value of driver time saved. (Recall from p. 38 that 2001 congestion cost
~$70 billion per year.) The net cost of our partial IHS portfolio is therefore
at worst zero and is probably strongly negative. However, the lack of reliable cost figures doesn’t matter because we’ve already credited IHS with
helping to offset rebound (p. 41), so it’s not in our oil-efficiency supply
curves.380 If it were, it would probably reduce the average cost per saved
barrel.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation
Other civilian highway and off-road vehicles
Technical Annex, Ch. 9, discusses the minor oil use by other highway and offroad vehicles, such as agricultural and construction equipment. Urban transit buses are not discussed here but on p. 120, in the context of fuel-switching, because, we’ll argue, by 2025 they can and should be 100% converted
to using saved natural gas (partly to improve urban air quality), saving an
eighth of a million bbl/d with a payback time of about half of a bus’s typical
operating life. Bus efficiency will thus save natural gas rather than oil.
Trains
Technical Annex, Ch. 10, based on industry and government studies,
describes Conventional Wisdom technologies to save 13% of trains’ 2025 fuel
use. Full implementation would save 0.04 Mbbl/d in 2025 at 14¢/gal of
diesel or $2.9/bbl crude. State of the Art technologies would save 30%, or
0.1 Mbbl/d in 2025, at 26¢/gal diesel or $7.8/bbl crude. These are on top
of mostly technological 1%/y improvements—28% during 2000–2025—
assumed by EIA. The Swiss railways, which have very active R&D and are
already rather efficient, foresee even larger potential savings (up to 60%)
from integrating new propulsion concepts (up to 30%), lightweighting
(up to 20%), cutting drag and friction (up to 10%), and optimizing operations.381 A 60% saving in Switzerland is a 66% saving against the 2000 U.S.
freight railways’ baseline, which is 17% more energy-intensive to start
with.382 The forecast Swiss savings are thus half again as big as our U.S.
State of the Art savings for 2025, suggesting ours are conservative. The more
advanced propulsion concepts, exceeding 310 mph, could also displace the
least efficient air travel.
Ships
Marine transportation offers potential to save 31% of its 2025 residual oil
use, or 0.2 Mbbl/d, at a cost of $0.12/gal of residual oil using CW technologies, or 56% (0.4 Mbbl/d) at a cost of $0.23/gal with SOA. The portfolio includes improved hull shape and materials, larger ships, drag reductions, hotel-load savings, and better engines and propulsors, plus a little
logistics and routing improvement. Details are in Technical Annex, Ch. 11.
Airplanes
Civilian air transport of passengers and freight is projected by EIA to use
86.5% of aviation fuel in 2025; the rest fuels military platforms. We focus
here on civilian airplanes, and later (pp. 84–93) apply their lessons to civilian-like parts of the military fleet. We consider here only kerosene-fueled
jet airplanes; potential liquid-hydrogen-fueled versions are discussed
below at note 916, p. 239.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
We count urban
transit buses’
efficiency gains later
under natural gas,
because they’ll have
been converted to use it.
Our State of the Art
railway energy saving
is only two-thirds
of what the Swiss
National Railway
envisages.
381. Jochem 2004, p. 25.
In the 1990s, for example,
the Danish Railway (SB)
developed the Copenhagen
S-train, with 46% lower
weight per seat than the
1986 model.
382. Swiss freight railways
in 2000 used 232 kJ/tonnekm (SSB 2003). The corresponding U.S. figure was
352 BTU/ton-mi or 257
kJ/tonne-km (ORNL 2003,
p. 9-10). However, we don’t
know how much adjustment,
if any, is required for the
freight dominance of U.S.
railways’ high-load-factor
but one-way coal-hauling.
Proven technologies
can cheaply save
more than half of the
nearly 3% of 2025 oil
used by ships.
The industry agrees
that airliners in 2025
can cost-effectively
save half to twothirds of forecasted
fuel use by the
national fleet, or up to
5 percent of U.S. oil
use, while improving
comfort, safety,
and cost.
79
Saving Oil
Option 1. Efficient use of oil: Transportation: Airplanes
Airplanes’ efficiency in flight
depends383 on engine efficiency,
weight, aerodynamic lift/drag
ratio, layout, load factor, and electric power generation and accessories.384
Airplanes certified during 1960–2000 showed a 70% decrease in “block
fuel use” (gate-to-gate fuel used in defined and idealized “normal” operations), coming about half from better engines and half from better airframes.385 This impressive progress decelerated somewhat in the 1990s,
mainly because the easiest performance gains were already achieved, airlines’ financial weakness retarded development of new models, and slow
adoption of advanced composites caused weight savings to lag behind
engine and aerodynamic improvements. Yet even more than for cars,
weight is the most critical factor, because taking one pound out of a midsize airplane typically saves ~124 pounds of fuel every year,386 worth more
than $200 over 30 years. The composite fraction of structural weight—
only about 3% in a 767 and 9% in a 777—will exceed 50% in the nextgeneration midsize 7E7 and (according to press reports) in the superjumbo A380, as noted on p. 56. The historic ~3.3%/y drop in energy intensity
may be resumed if such innovations are rapidly taken into the global
fleet. Most forecasters doubt this because of the industry’s financial weakness, but policy (pp. 154–159) could change that. A leading U.S. team
expects a 1.2–2.2%/y drop in energy intensity to 2025 (1.0–2.0%/y without higher load factors); 386a Airbus expects 2%/y (Fig. 24); DASA387 opined
that even 3.5%/y is theoretically possible.388
Taking one pound out of a midsize airplane typically saves ~124 pounds
of fuel every year, worth more than $200 over 30 years.
383. Greene 2004.
384. Such auxiliary loads as
air-handling, space-conditioning, lighting, cooking,
and electronics have not yet
received the systematic and
up-to-date attention they
deserve. RMI’s worldwide
observations and discussions with a major aircraft
manufacturer suggest that
redesigning auxiliary loads
could profitably save up to
2% of fuel use in, say, a 7E7
(which will be several times
as electricity-intensive as its
predecessors because it
does electrically many
things previously done with
fluids). This can save weight
and fuel carriage, while
improving passengers’ comfort, health, and safety.
To our knowledge, no manufacturer has yet systematically applied in these uses
the efficiency innovations
practiced in today’s most
efficient buildings. Absent a
rigorous determination of
which of the improvements
aren’t yet counted in the
7E7 design, this topic is
conservatively omitted from
our efficiency analysis.
A fully loaded 7E7’s expected ~97 passenger-miles per gallon is impressive—like a State of the Art car, only going ten times as fast—but is far
from the ultimate. Our bottom-up calculation estimates the effects and
costs of established techniques for reducing weight and drag, engine losses, and other minor terms. Fortunately, such gains’ parametric effects
have been extensively analyzed, so we have relied on industry forecasts
to 2025. We separately assess small, medium, and large jets (comparable
to 737-800, 777-200ER, and 747-400) in EIA’s 2025 fleet mix. Regional airplanes today (7% of civilian aviation fuel use, 17% in 2025) are ~40–60%
less fuel-efficient than long-haul jets, partly because their short stages,
385. Compared with
a DC8–21: Lee 2000.
386. D. Daggett (Boeing), personal communication, 29 August 2003. A similar analysis for a different, double-aisle airplane yielded a similar result (111 lb/y: D.
Daggett, personal communication, 28 May 2004). At EIA’s 2025 price of 80.7¢/gal (2000 $) and our 5%/y real discount rate, the 124:1 mass-decompounding ratio
yields a 30-year present value of $228 worth of fuel for each pound of avoided weight. Alternatively, nominal industry assumptions imply that saving a pound
could be worth more than that in profit (tens of times that much in revenue) if it allowed a pound of extra payload to be carried all the time at market prices
(R.L. Garwin, personal communication, 26 May 2004).
386a. Lee et al. 2001.
387. DASA, the Munich-based DaimlerChrysler Aerospace AG, merged in 2000 with Aerospatiale Matra and Casa to form EADS (European Aeronautic
Defence and Space Company), the parent of Airbus.
388. Faaß 2001, slide 11. Unfortunately the DASA study graphed there is not publicly available, so we have not graphed its ~0.9 MJ/RP-km (megajoule-perrevenue-passenger-kilometer) potential (20% above RMI’s State of the Art best case shown in Fig. 25).
80
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Airplanes
short-range
projected
13
long-range
U.S. fleet average
(at actual load factor)
6
(2) medium range (737-400 )
U.S. fleet average
(at 2000 load factor = 0.72)
5
DC9-10
(3) regional jet
DC9-30
B727-200
3
DC9-40
2
B747-200/300
B737-100/200
DC10-30
2025
2020
2015
1995
1990
1985
1980
1975
0
RMI CW (3)
(2) Delft study
RMI SOA (3)
(2)
(1)
RMI CW
(1)Delft study NRC
B7E7
(1),(2)
(1),(2)
Airbus
(based on
Greene (1),(2)
(1),(2)
(2) Delft
B767-300–20%)
(1) study
Lee et al. (2) RMI SOA (1),(2)
2005
1
A300-600
MD80
MD-11
B777
B767-200
B747-400
B737-300
B757-200
A320-100/200
B767-300
B737-700/800
B737-400
2000
DC10-10
1970
25
EIA projected commercial
airplane fleet average
2010
4
(1) long-range (747-400 )
50
100
200
energy intensity (revenue-passenger-mile/gal.)
actual
7
1965
energy efficiency (MJ/revenue passenger-km)
Figure 25: Historic and projected airplane energy intensities, 1955–2025
This RMI graph shows the evolution of more efficient kerosene-fueled airplanes. EIA assumes efficiency gains will slow by three-fifths,
making the 2025 fleet two to three times less efficient than the most efficient airplanes then expected to be available and cost-effective—a bigger percentage gap than today’s. Historical airplane fuel-intensities are from actual quarterly data for 1998–2003, showing
the mean and one-standard-deviation error bar.389 (See Technical Annex, Ch. 12). The discontinuity between historical 390 and projected 391
fleet averages is due to differences in fleet definitions (including cargo, charter planes, and the fraction of international fuels accounted
for). EIA projects fleet and new airplane efficiency to increase 0.8%/y and 0.4%/y (1.1%/y and 0.6%/y including the 72%→76% rise in
load factor) during 2000–2025. EIA’s technology model adopts an assumption of weight-saving materials (99% adoption by 2017, raising
seat-mpg 15% vs. 1990) and ultra-high bypass engines (68% adoption by 2025, raising seat-mpg 10% vs. 1990), but EIA considers its
other four technologies (propfan, thermodynamics, hybrid laminar flow, and advanced aerodynamics) not cost-effective at $0.81/gal
(2000 $) in 2025. In contrast, RMI believes a much more powerful technology portfolio to be competitive. Airplane projections are based
on operational data using EIA’s projected load factor for the year in which each study is shown. Operational data are analogous to onroad rather than EPA laboratory or adjusted fuel intensity for light vehicles. For example, on an idealized 2000-man mission, 7E7 uses
only 0.875 MJ/RP-km, vs. 1.6 MJ/RP-km derived from 767-300–209 and graphed here.392 The difference, due to operational imperfections
and age degradation, can be reduced by better practices and technologies.393 Boeing’s idealized efficiency data are in Technical Annex,
Ch. 12. These are additional to the airplane-technology-based difference between the fleet-average and airplane projections.
The Airbus,394 EIA,395 Greene,396 and NRC 397 projections shown represent fleet averages for the given year, while the Lee et al.,398 Delft,399
and RMI projections are airplane-specific.
year, or year of introduction for airplanes
Source: RMI analysis; see caption and notes for sources.
389. BTS 2004.
390. ORNL 2003, Table 9.2.
391. EIA 2004.
392. M. Mirza, Boeing Economic Analysis Department,
personal communication, 11 June 2004, based on a
standardized 2,000-nautical-mile trip at the 2000 average
load factor of 0.72; Boeing, undated (downloaded 13
August 2004).
393. The difference between block and actual fuel use
in 1992 averaged ~18%, for complex but well-understood
reasons susceptible to considerable improvement
(Daggett et al. 1999).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
394. Faaß 2001.
395. EIA 2004.
396. Greene 1992.
397. CAT/NRC 1992.
Gains in airplane efficiency
are projected to keep on decelerating to 2025,
causing a two- to three-fold
“efficiency gap”
compared to the best models
then available.
398. Lee et al. 2001.
That paper’s Fig. 6 inspired Fig. 25 above.
399. Dings et al. 2000.
81
Saving Oil
Option 1. Efficient use of oil: Transportation: Airplanes
A 95%-advancedcomposite advanced
tactical fighter airframe design turned
out to be one-third
lighter but two-thirds
cheaper than its 72%metal predecessor.
especially those less than 600 miles, decrease their ratio of cruise flight to
less efficient air and ground modes. But big efficiency gains can be had in
all sizes, and regional jets may displace some larger airplanes.400
400. Lee et al. 2004.
401. Pole 2003.
402. Typically these use
lighter, stronger materials
at higher temperatures
and pressures. Only about
a fourth to a third of fuel
energy typically moves the
aircraft; the rest is discharged as the engine’s
waste heat (Lee et al. 2004).
This isn’t for lack of trying—modern engines use
single-crystal titanium
blades, and peak temperatures are about one-third
that of the surface of the
sun—but remarkably,
further 10–30% improvements are in store.
Improving different parts of the integrated airplane system tends to be
synergistic. Boeing says 401 that as composites make 7E7 lighter but more
spacious (Fig. 25a), its one-fifth drop in fuel intensity, vs. a comparably
sized 767–300, will come 40% from better engines,402 30% from better airframe materials, aerodynamics, and systems architecture, and the other
30% from “cycling the design”: more efficient engines and better lift/drag
ratio reduces fuel loading, thus optimizing requirements for the wing
and landing gear, thus saving even more weight and fuel, and so on.
This “design spiral” is like that of ultralight cars, only more so, because
the vehicle is carried by wings rather than wheels. And as with composite
cars, composite airplane structures reduce weight while adding other
benefits, including lower cost. Airbus’s proposed future composite fuselage aims to cut basic fuselage weight by 30% and cost by 40% while
eliminating structural fatigue and corrosion (both maintenance concerns)
and improving passenger comfort.403 Moreover, much of the 7E7 and A380
composite is not advanced polymer composite but “glare”—compositereinforced metal—leaving even more room for further weight savings.
Some of the efficiency typically projected in design studies tends to get
lost in translation into an actual commercial airplane. But conversely,
the industry studies we’ve relied upon—which come out close to our
own findings, as Fig. 25 shows—don’t include all significant options.
For example, a 1994–96 military project suggests that higher advancedcomposite fractions could save even more weight and cost than normally
assumed: a 95%-composite advanced tactical fighter airframe design
turned out to be one-third lighter but two-thirds cheaper than its 72%metal predecessor.404 This approach would be especially advantageous in
a Blended Wing Body passenger aircraft (Fig. 25c)—an advanced design
we favor for State of the Art, hard to make with Conventional Wisdom’s
tube-and-wing metal-forming techniques, but ideally suited to molded
advanced composites.
In addition, improvements in the air transportation system and logistics,
chiefly from information technology, are expected to save an additional
5–10% of system fuel at negative cost, both in the air and on the ground.
403. Kupke & Kolax 2004.
404. This DARPA-funded Integrated Technology for Affordability (IATA) design project was conducted in 1994–96 at the Lockheed-Martin Skunk Works® with
support from Alliant Techsystems, Dow-UTL, and AECL, and briefed on 20 September 2000 to the Defense Science Board panel “Improving Fuel Efficiency of
Weapons Platforms” by D.F. Taggart, then chief engineer of Hypercar, Inc. and previously leader of the IATA effort at the Skunk Works. Although too revolutionary at the time to find a customer, it clearly showed that the same radical simplifications of structure and manufacturing (including fastenerless self-fixturing assembly) that an advanced-composites-dominated design has brought to cars (Lovins & Cramer 2004) should also work for aircraft, both military and
civilian. In this instance, compared with the JAST 140 conventional design, the IATA wing/body design was 33% lighter, had a lower recurring production
cost (by 77% at T1, 65% at T100, 73% at T250), and had a 48% lower nonrecurring production cost and orders of magnitude fewer parts.
82
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Airplanes
Figure 26: Next-generation airplanes. L–R: Interior vision of Boeing’s mostly-composite long-range 7E7
Dreamliner, to enter service in 2008 with 15–20% fuel saving at no greater cost; Northrop flying-wing B-2
bomber, showing how the low-signature engines of a Blended-Wing-Body design can be packed inside;
artist’s conception of a civilian BWB airplane without this potentially efficiency-improving feature;405
and an artist’s conception of a civilian BWB with ultra-efficient engine technology, embedded wing
propulsion, and boundary-layer ingestion engine inlets with active flow control.
Many air-traffic-management innovations are also expected to increase
safety despite higher traffic volume. All these improvements together could
save 45% of EIA’s projected 2025 airplane fuel intensity at an average cost
of 46¢/gal of saved aviation fuel using State of the Art technologies, or 21%
at 67¢/gal with Conventional Wisdom options. Details are in Technical Annex,
Ch. 12. Fitting the pattern that emerged in several analyses of road vehicles,
the more advanced State of the Art technologies tend to save more energy at
comparable or lower cost because they use more integrative
platform designs that capture more synergies, achieving expanding rather
than diminishing returns to investments in saving fuel.
Just as today’s car fleet is only half as efficient as a Prius, the U.S. passenger jet fleet is only about two-thirds as efficient as a 777. By 2025, the best
new planes could get 2.9× the seat-miles per gallon of today’s fleet average, or 2.3× more than EIA’s projection, leaving a huge overhang of
unbought efficiency. The barriers to efficient aircraft are not necessarily
only technological, but partly the understandable conservatism of the
manufacturers and regulators, and chiefly, as with light vehicles, arise
from business dynamics. Although Boeing appears to be pricing the 20%more-efficient 7E7 at little or no—even negative (note 663, p. 157)—extra
cost, most airlines can’t afford rapid fleet replacements, and airframes fly
for 20–50 years, usually as hand-me-downs to small airlines, cargo carriers, and developing countries. Airlines are a great industry but a bad
business: as Warren Buffett famously calculated,406 U.S. airlines have collectively earned zero cumulative net profit since the Wright brothers
(1903–2003), to which Southwest Airlines’ Herb Kelleher added a decade
ago, “If the Wright brothers were alive today, Wilbur would have to fire
Orville to reduce costs.”407 Bankrupt or capital-strapped operators can’t
afford to buy the best that technological innovators can make, but adoption can be greatly accelerated, as we’ll propose on pp. 154–159.
Over the next two
decades, advanced
airplanes can save
45% of projected fuel
at 46¢ a gallon.
Many of the most
sophisticated technologies will cost
less than small,
incremental savings.
The U.S. airline
industry has run a
cumulative net
financial loss since
the Wright brothers,
and remains
handicapped
by fuel inefficiency.
405. Holmes 2002. The B-2 achieved a very low radar signature at the expense of fuel economy because of engine airflow
distortion. A commercial BWB airplane’s buried engines could save fuel, due to smaller exposed surface area,
if they used novel active airflow control devices to avoid either a sharp S-bend or a long but heavy and high-drag S-duct.
406. As of 1995, after which the industry made a little money but then lost far more.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
407. Jones 1994.
83
Saving Oil
Option 1. Efficient use of oil: Transportation
Huge efficiency
gains by the Pentagon—
the world’s largest
oil buyer—could make
military force more
effective, improve force
protection, and cut
defense costs by
more than $10 billion
a year.
Military vehicles
DoD buys five billion
gallons a year—
enough to drive
every U.S. car coastto-coast every four
years.
Fighting an intensive
Gulf war uses oil
at about the same
rate at which the
U.S. imports oil from
Kuwait.
Of the gross tonnage
moved when
the Army deploys,
~70% is fuel.
The Army directly
uses only ~$0.2 billion
worth of fuel
a year, but pays
~$3.2 billion a year
to maintain
20,000 active-duty
and 40,000 reserve
personnel to move it.
Military use of oil is only 1.6% of 2000 and 1.5% of projected 2025 national
total use (Fig. 6). However, its implications for security and the macroeconomy are disproportionate. In particular, its ability to accelerate a massive technological shift in the civilian economy, as it did with microchips,
merits extended discussion here.
War is among the most energy-intensive human activities, consuming about
two-fifths of total U.S. energy in 1941–45.408 The ~$0.4-trillion/y, 3-millionperson Department of Defense (DoD)—reportedly the nation’s oldest and
largest organization—operates 600,000 structures on 30 million acres in 6,000
locations in 146 countries; 550 public utility systems; hundreds of thousands
of land vehicles; hundreds of ocean-going vessels; and more than 20,000 aircraft. Mainly to fuel these platforms, DoD spends upwards of $5 billion a
year on energy. It is the largest U.S. buyer of energy, using 1.1% of national
and 85% of government energy in 2002. DoD is probably also the world’s
largest oil buyer—five billion gallons a year, enough to drive every U.S. car
coast-to-coast every four years. If the Pentagon were a country, it would
rank in the top third of energy users worldwide.
Out of every eight barrels that fueled the Allies’ victory over oil-starved
Nazi Germany and Japan in World War II, seven came from American
wells.409 That would be impossible today, when warfare is ~15 times as
oil-intensive410 and most U.S. oil is imported. In 2001, U.S. warplanes over
Iraq were fueled partly with oil from Iraq, then the nation’s sixth largest
supplier.411 In Operations Desert Shield/Desert Storm (1990–91), 75% of oil
was sourced in-theater from cooperative Gulf neighbors; Saudi Arabia
provided 21 million gallons a day. In the 2002 Afghan campaign, 95% of
the Defense Logistics Agency’s half-billion gallons of fuel was trucked in
from Pakistan, Uzbekhistan, and elsewhere in Central Asia, including
Russian jet fuel left over from the 1980s Soviet/Afghan war.412
The fuel logistics burden
Fighting an intensive Gulf war uses oil, mostly for aircraft,413 at about
the same rate at which the U.S. imports oil from Kuwait.414 That use is tiny
in world oil markets: two wars in Iraq plus one in Afghanistan used a
total of only ~103 Mbbl—1.5% as much as the U.S. uses in a year, or 5%
of the two billion barrels Saddam Hussein’s forces torched in Kuwait.415
408. War is both expensive—in 2000 $, an estimated $250 billion for World War I, $2.75 trillion for World War II, and $450 billion for the Vietnam War—and
more energy-intensive than the civilian economy’s average, by factors estimated at very roughly 1.5, 2, and 3 respectively (Smil 2004).
409. Painter 2002; Klare 2004.
410. Comparing, for example, the 1991 Gulf War with the liberation of Europe at the end of World War II: Copulos 2004.
411. Ebel 2002.
84
412–415. See next page.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil: Transportation: Military vehicles: The fuel logistics burden
Still, it’s an immense logistical strain to deliver that much oil that quickly
into theater, often to austere sites, and especially into thirsty platforms
that mustn’t run dry. Fuel logistics, as much as anything, prevents
America’s most lethally effective forces from being rapidly deployable
and its most rapidly deployable forces from assuredly winning.416
Weapons themselves are seldom417 very energy-consumptive—they merely
focus energy into a specific zone of destruction for an extremely short
time—but the platforms used to carry weapons systems tend to be extremely energy-intensive and must be fueled by a large, complex, globe-girdling,
and (in its own right) energy-intensive logistics chain. Of the gross tonnage
moved when the Army deploys, ~70% is fuel, and transporting that fuel
to the bases or depots from which it’s redistributed costs about 8% as much
as the fuel itself.418 Over 60% of Air Force fuel is used not by fighters or
bombers but for airlifting people and materiel, including fuel.419
Military efficiency potential
The energy intensity of land, sea, and air platforms reflects four needs:
superior tactical performance (speed, maneuverability, endurance); carrying and delivering munitions; carrying armor and other forms of protection; and resisting or offsetting battle damage. All can be systematically
improved through better technologies, tactics, and operational concepts.
Within these four unique operational constraints, it is feasible, just as in
civilian vehicles, to reduce military platforms’ fuel use by methodically
applying technologies that cut weight, drag, idling, and auxiliary power
consumption and that raise the efficiency of converting fuel into motion
through engines, powertrains, and propulsors. In military as civilian vehicles, “The most important factor in reducing…demand for fuel… is reduc-
In 2001, U.S. warplanes
over Iraq were fueled
partly with oil from Iraq,
then the nation’s sixth
largest supplier.
Despite the special
requirements of tactical
performance, force
protection, munitions
carriage, and battle
damage resistance,
military platforms can
save oil in the same
ways as civilian ones—
by cutting weight,
drag, idling, auxiliary
power, and powertrain
losses—only the
savings are greater
because military
platforms are less
efficient to start with
and their fuel logistics
and its vulnerability
make savings far more
valuable.
412. Erwin 2002a; Cohen 2003. However, for the first two months of Operation Enduring Freedom, “major shortages of fuel plagued the force” despite half of
6,800 Air Force sorties, “flying the wings off airplanes,” being for fuel resupply, which Under Secretary of Defense Pete Aldridge described as “a terrible
way” to deliver fuel: Granger 2003.
413. In FY2002, 56% of all DoD energy use for both platforms and facilities, and 85% of the oil DoD contracted to procure, was aviation fuel (DESC 2003).
In the first seven months of 2002 Afghan operations, only 7% of fuel went to the Army (Cohen 2003).
414. In 1987, the Department of Energy estimated that DoD energy consumption could double or triple in a war (DOE 1987). Applying these factors to the
petroleum fraction (79%) of total FY2001 DoD energy consumption (0.774 quadrillion BTU/y—40% less than FY1987, due largely to force reductions) implies
that war could raise DoD’s oil use by ~0.3–0.6 Mbbl/d, consistent with the Defense Energy Support Center’s normal provision of 100 Mbbl/y or 0.27 Mbbl/d.
For so long as a wartime optempo (pace and intensity of operations) is sustained, that rise would equate to 12–24% of net imports from the Gulf, or to the
2000 rate of net imports from Kuwait (0.27 Mbbl/d) or Iraq (0.62 Mbbl/d). Actual fuel use in Desert Shield/Desert Storm was 1.9b bbl (0.25 Mbbl/d crude-equivalent); for the first 42 days of Operation Iraqi Freedom, at least 0.24 Mbbl/d (0.12 Mbbl/d over the first 84 days, 0.054 Mbbl/d during 19 Mar 2003 through 9 Feb
2004); and for the less intensive Operation Enduring Freedom (in and around Afghanistan), 0.050 Mbbl/d (1 Oct 2001–11 June 2003). By 9 Feb 2004, these four
operations had consumed a total of 4.5 billion gal, energy-equivalent to 103 Mbbl of crude oil—as much oil as the U.S. uses for all purposes every five days.
Sources: Cohen 2003; Erwin 2002a; Volk 2004; DESC 2003–04.
415. Laherrere 2004.
416. Logistics failures have defeated generals from Napoleon to Rommel (Van Creveld 1977).
417. The main exception is nuclear weapons because of their exotic materials and the huge energy inputs needed to enrich uranium and produce plutonium.
Smil (2004) estimates that at least 5% of all U.S. and Soviet commercial energy used during 1950–90 went into developing and producing these weapons and
their delivery systems—not implausible in view of their whole-system cost.
418. In FY2002 for all Services (DESC 2003), dividing delivery cost (p. 32) by value of petroleum products purchased (p. 19).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
419. DSB 2001.
85
Saving Oil
Option 1. Efficient use of oil: Transportation: Military vehicles: Military efficiency potential
Abrams tanks’ inefficient and mostly-idling
~1968 gas turbine
halves their fuel efficiency. This delayed
the invasion of Iraq
by more than a month
to stockpile
the extra fuel.
ing the weight of…vehicles.”420 But there are two key differences: in military platforms, fuel efficiency is immensely more valuable—not only in dollars
but in lives, and potentially even in the margin between victory and
defeat—and most military structures and engines use decades-old technologies,
offering even more scope for improvement. Box 11 (pp. 86–90) illustrates
these valuable opportunities to save oil in DoD’s existing inventory.
The Air Force spent
84 percent of its fueldelivery cost on the
6 percent of its gallons
that were delivered
in midair.
Beyond the retrofits summarized in Box 11, there’s even greater potential
scope for more fight with less fuel at lower cost. Technical Annex, Ch. 13,
presents an initial sketch of how the findings of a seminal 2001 Defense
Science Board report could apply to the estimated 2025 fleet of land, sea,
and air platforms. The result: 61% Conventional Wisdom and 66% State of
the Art reductions in petroleum-based fuel use by all DoD land, sea, and
air platforms, not counting most of the fuel saved by transporting less
metal and fuel.
420. NAS/NRC/CETS 1999,
p. 148.
Such radical State of the Art savings may surprise those unaware that
DoD’s official Army After Next goals for 2020 include 75% battlefield fuel
421–441. See Box 11.
(continued on p. 91)
11: Saving oil in existing military platforms
The nearly 70-ton M1A2 Abrams main battle
tank—the outstanding fighting machine of U.S.
(and Saudi, Kuwaiti, and Egyptian) armored
forces—is propelled at up to 42 mph on- or
30 mph off-road by a 1,500-hp gas turbine,421 and
averages around 0.3–0.6 mpg.422 Its ~20–40 ton421. Rather than the diesel engine used in all other Army heavy land
platforms. Diesels are historically heavier and bulkier, but may not be
any more, and use less fuel, which has its own weight and bulk
(Erwin 2000).
422. Its nominal fuel consumption (from a 505-gal fuel capacity)
ranges from 60 gal/h gross-country to 30+ “while operating at a tactical ideal” to 10 in basic idle mode; nominal consumption averages
~37.5 gal/h; nominal cruising range is 265 mi/505 gal = 0.52 mpg
(GlobalSecurity.org 2004). However, worse values, such as crosscountry ~0.3 mpg and idling ~16 gal/h, have been reported
(Periscope1.com, undated).
423. The late-1960s-technology AGT 1500 gas-turbine engine, out of
production since 1992, was supposed to be replaced starting in FY03
or FY04 by the LV100 engine, which was predicted to increase range
by up to 26%, halve idling fuel use, and quintuple the AGT 1500’s
notably poor reliability (Geae.com, undated). The AGT 1500’s rated
nominal fuel intensity was an unimpressive 0.45 lb/shp-h (poundsper-shaft-horsepower-hour) (Turbokart.com, undated) or ~31%—
much worse at part-load. However, cancellation of Crusader, a
heavy mobile artillery system that was also to use the LV100, raised
(Continued next page.)
86
mpg is surprisingly close to the ~42 ton-mpg
of today’s average new light vehicle; the tank
simply weighs ~34 times as much, half for armor.
But there’s more to be done than improving its
~1968 gas turbine: 423 for ~73% of its operating
hours, Abrams idles that ~1,100-kW gas turbine
at less than 1% efficiency to run a ~5-kW “hotel
load”—ventilation, lights, cooling, and electronics. This, coupled with its inherent engine inefficiency, cuts Abrams ’s average fuel efficiency
about in half,424 requiring extra fuel whose stockpiling for the Gulf War delayed the ground forces’
readiness to fight by more than a month.425
This is one of many striking examples that
emerged in a 1991–2001 Defense Science Board
(DSB) task force on which one of us (ABL)
served under the chairmanship of former astronaut, NASA director, naval aviator, and Vice
Admiral (Ret.) Richard Truly. This widely noticed
(continued on next page)
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Box 11: Saving oil in existing military platforms (continued)
unclassified report 426 covering all Services’ land,
sea, and air platforms found more than 100
effective fuel-saving technologies that could
save much, perhaps most, of their fuel while
never impairing and often improving warfighting
capability.
The DSB panel examined why a capable meritocracy with more wants than funds hadn’t
already achieved such large and lucrative savings. The institutional reasons are complex, but
one of the biggest is false price signals due to
lack of activity-based costing. When weapons
platforms are designed and bought, the cost of
their fuel has been assumed to equal the average wholesale price charged by the Defense
Energy Support Center (DESC), historically
around a dollar per gallon. Logistics—moving
stuff around—occupies roughly a third of DOD’s
budget and half its personnel. But when designing and buying platforms, logistics is considered
free, because the cost-benefit analyses used to
justify investments in efficiency are based on
the DESC price for fuel, excluding the cost of
delivering it . This understates delivered fuel
cost by a factor of at least three, perhaps
several times three (Fig. 27 427)—even by tens or
hundreds in some specific cases.
(fn 424 continued)
that engine’s unit cost, leaving Abrams re-engining in limbo (Erwin 2003).
There may also be other alternatives, such as the experimental Semi
Closed Cycle Compact Turbine Engine, expected to more than double
AGT 1500’s efficiency yet occupy far less volume (Salyer 1999, slide 33).
Abrams illustrates why. Since gas turbines,
inefficient at best, become extremely so at low
loads,428 a small APU matched to the small, fairly
steady 5-kW “hotel load” would save 96% of the
fuel wasted in idling the huge gas turbine.
Abrams was designed with no APU on the
assumption that its fuel would cost ~$1/gal with
zero delivery cost. But to keep up with a rapid
armored advance that outruns resupply trucks,
bladders of fuel may have to be slung beneath
cargo helicopters and leapfrogged 400+ km into
theater in a three-stage relay (eight helicopters
at the front end to get one to its destination),
consuming most of the fuel in order to carry the
rest. Delivery cost can then rise to an eye-popping $600 a gallon, becoming astronomical
beyond 400 km.429 Even delivery by land to the
Forward Edge of the Battle Area (FEBA) costs
~$30/gal. If Abrams ’s designers had considered
fuel transport cost, they’d probably have
designed the tank very differently. Yet misled by
false cost signals, they didn’t leave room under
armor for an APU. The DSB panel suggested a
Russian-style pragmatic improvisation: buy a
Honda genset at Home Depot and strap it onto
the back of the tank. Most of the time, nobody is
shooting at the tank, and the genset will save
nearly half Abrams ’s fuel. If the genset ever gets
shot away, Abrams is no worse off than it is right
now. (The Army has instead begun developing an
APU to squeeze into scant under-armor spaces.)
(continued on next page)
424. Using M1A1 data (79% of the 2000 M1 fleet), the fleet-average
utilization is 205 h/y or 411 mi/y, totaling 55 h/y mobility and 150 h/y
idle at 12 gal/h, of which the Army believes 98 h/y or 65% could be displaced by a 0.5 gal/h APU, saving 46% of the tank’s total fuel consumption
(Moran 2000, slide 14). (The Army calculates average savings of only 32% for the M1A2 because it assumes the APU uses 4 gal/h. We suspect more
advanced APUs could use less fuel and run longer.) Potential APU savings are largely additional to those of making the main engine more efficient,
which would easily bring the Army’s APU-saving estimate (a fleet-weighted average of 43%) to well above 50%.
425. Salyer 1999, slide 8, assuming a corresponding reduction in the fuel and infrastructure inserted.
426. DSB 2001. Summarized by Truly (2001); in Book 2002; and in Ginsburg 2001.
427. DSB 2001; the previously unassembled data used to compile this chart come respectively from pp. 39, 4 and 20, and 17, adding only the Navy’s
information about the split of oil delivery modes between oiler (70%) and pierside (30%) at respective costs of $0.64/gal and $0.05/gal (Alan Roberts
[OSD], personal communication, 3 April 2001).
428. To produce power without stalling (being aerodynamic devices), gas turbines must keep spinning rapidly regardless of their load; they cannot simply run very slowly like an internal-combustion engine.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
87
Saving Oil
Box 11: Saving oil in existing military platforms (continued)
Fuel logistics hamstrings the Army
Fuel-wasting design doesn’t just cost money;
it weakens warfighting. Each tank is trailed by
several lumbering 5,000-gallon tankers. An
armored division may use 20, perhaps even 40,
times as many daily tons of fuel as it does of
munitions—around 600,000 gallons a day. Of the
Army’s top ten battlefield fuel guzzlers, only #5
(Abrams tanks) and #10 (Apache helicopters)
are combat vehicles; three of the top four are
trucks,430 some carrying fuel—not unlike the sixmule Civil War wagons that hauled roughly a
ton of materiel, half of it feed.431 Some 55% of the
fuel the Army takes to the battlefield doesn’t
even get to front-line combat units, but is consumed by echelons above corps and by rear
units.432 This big logistical tail slows deployment,
constrains maneuver, and requires a lot of
equipment and people. The Army directly uses
only ~$0.2 billion worth of fuel a year, but pays
~$3.2 billion a year to maintain 20,000 active-duty
personnel (at ~$100,000/y) and 40,000 reserve
personnel (at ~$30,000/y) to move that fuel.
(continued on next page)
Figure 27: Uncounted costs of delivering fuel to weapons platforms
Summary of Defense Science Board panel’s 2001 findings of a severalfold difference between the undelivered “wholesale”
fuel cost assumed when requiring, designing, and procuring weapons platforms (blue) and the “retail” price they actually pay in
peacetime (magenta). Combat delivery to the platform can cost far more, and even in peacetime, the magenta bars omit many
large infrastructure and staff costs (p. 267).
DESC direct fuel cost @ $0.89/gal (FY99, 2000 $),
assumed when designing platforms
delivered fuel cost for fueling platforms
(difference is partial peacetime fuel logistics cost)
“logistics gap”
billion 2000 $/y
15
10
5
0
Army
Navy
Air Force
total DoD
Fuel used to deploy Army assets via Navy or Air Force lift is
ascribed to those Services; the size of these allocations is
unknown, but over 60% of Air Force fuel is used for lift. Some Air
Force refueling (~16% in Operation Desert Storm) serves Navy and
Marine Corps aircraft. Army and Navy data shown are for 1997; Air
Force, 1999; Navy includes Marine Corps. Major omissions from the
magenta bars include: Army's oil-related equipment and facilities
and their ownership cost; Army's fuel delivery into platforms, which
can cost hundreds of times the blue bar; Navy's purchase of new
oilers; and Air Force's acquisition of ~100 new aerial tankers (now
proposed, for tens of billions of dollars). At the FY02 DESC average
fuel cost of $1.29/gal (FY02, 2000 $), the delivered fuel price probably
rose to ~$12–14b. The direct fuel cost, shown as $3.7b at the FY99
DESC fuel charge, has since risen above $6b in FY2004–05 (2004 $)
due mainly to higher volumes delivered (the DESC FY04 fuel cost
was $0.91/gal, 2004 $). Delivery not just into theater but into the platform can become enormously costlier in combat: for land forces in
Iraq in 2003, the fully burdened logistics cost reportedly averaged
around $130/gal (2003 $), or about a hundred times the DESC direct
fuel price to which the platforms were designed.
Source: RMI analysis from DSB 2001; see note 427, p. 87.
429. Salyer 1999, slide 7.
430. For Army battlefield units broadly: Hunt 2003. Four of the Army’s top ten battlefield users are trucks: heavy linehaul, medium tactical, heavyequipment transporter, and HMMWV (“Humvee”). The Army’s quarter-million trucks, driving 823 million miles a year, are sometimes said to be the
world’s largest fleet: Higgins 2002; Gorsich 2000.
431. Hoeper 1999. This is not a new problem: NRC (1999), at p. 67, quotes Sun Tze’s The Art of War, ~100 BC, as saying: “Transportation of provisions
itself consumes 20 times the amount transported.” This appears to be a reference to 2:14, where Sun Tze notes that foraging off the enemy is
always to be preferred, because one cartload of his provisions is equivalent to 20 of your own (since you haven’t borne its logistical burden).
A summary of military logistics history (Gabriel & Metz 1992, Ch. 3) suggests that Alexander the Great’s triumphs were importantly based on a logistics-based early Revolution in Military Affairs.
432. Erwin 2002. The article notes a TACOM finding that if an Army Brigade Combat Team used hybrid vehicles, their range would increase by 180
miles, and each 100 miles would reduce the brigade’s fuel needs by 4,000 gallons, nearly a full tanker-truckload.
88
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Box 11: Saving oil in existing military platforms (continued)
Abrams tanks idle their main engine ~73% of the time, at less than 1% efficiency, to run a small “hotel load”
(limiting their “silent watch” capability to about a half-hour) because their fuel delivery was assumed to be free.
On the battlefield, rapid fuel delivery can cost $600 a gallon.
This logistics personnel cost thus multiplies the
$0.2b direct fuel cost by 16, plus further pyramiding gear and people—and represents a
timely opportunity to deploy a whole division of
personnel from tail to tooth. In a specifically
simulated example, the Army After Next equivalent of today’s heavy divisions would save not
only ~$0.33 million/d in battlefield fuel, but also
would eliminate the need for 9,276 oil, maintenance, and other personnel costing eight times
as much—about $2.5 million/d—not counting
additional personnel and equipment savings
stretching back to the homeland.433 By 2025, an
estimated science and technology investment
totaling on the order of $2 billion would be saving ~$1.2 billion each year, and rising fast, just
on oil-related personnel and infrastructure.434
Unarmored fuel carriers are also vulnerable.
Attacks on in-theater and rear logistics assets
can make a fuel-hungry combat system grind to
a halt—as it did, to great upset, in a recent
wargame that was stopped because this tactic
was considered unfair. (The U.S. is fortunate that
the distinctly unsporting Saddam Hussein didn’t
use it more.) Yet the warfighting benefits of fuel
433. Salyer 2000, slide 19.
434. Salyer 2000, slide 25.
435. The short-run marginal cost components for FY99, in 2000 $, are
$17.03 air-to-air cost, $0.22/gal delivery into the tanker on the
ground, and $0.89/gal DESC fuel charge delivered to the airbase
(DSB 2001).
436. Loitering is important for killing time-sensitive, high-value targets using near-real-time intelligence. Of course, better engines’
performance can be exploited by any combination of range, loiter,
payload, or other performance metrics. For example, better
(IHPTEP) engines, the DSB panel was told, could let Naval combat
air patrols carry more payload (due to 36% lower takeoff gross
weight) or stay out longer (44% lower fuel burn at constant mission).
In antisubmarine helicopters, better engines could increase radius
by 430% at constant payload and loiter time, or increase payload by
80% at constant radius and loiter. Such numbers represent very
valuable force multipliers, amounting to free new “virtual” ships
and aircraft.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
economy—in deployability, agility, range, speed,
reliability, autonomy, dominant maneuver in an
extended battlespace, etc.—are as invisible to
platform designers as logistics cost: none of
these advantages is yet valued by DoD buyers,
so inefficient platforms keep getting bought,
wasting oil and money and hobbling warfighters.
Air Force: gas stations in the sky
What about the Air Force? The venerable B-52H
bombers now being flown by the children of
their original pilots (and structurally capable of
flying until at least 2037) have inefficient, lowbypass, 1950s-technology engines. By 2010,
those could be refitted to today’s commercial
engines, using 33% less fuel to fly 46% farther.
On its usual long-distance and long-loiter-time
missions, B-52H typically needs midair refueling
costing ~$18.1 a gallon (2000 $).435 (That cost
doesn’t include at least 55 tankers the Air Force
will soon need to replace for more than $10 billion.) Thus the Air Force in FY1999 paid $1.8 billion for two billion gallons of fuel, but delivering
that fuel into the aircraft added another $2.6 billion, so the actual delivered fuel bill was $4.4
billion: the Air Force spent 84 percent of its fueldelivery cost on the 6 percent of its gallons that
were delivered in midair. Counting that delivery
cost, re-engining B-52H would repay twice its
cost; greatly reduce or even eliminate midair
refueling; free up scarce tankers and ramp
space for other missions; and enable bombers
to loiter for hours over their targets.436 A Defense
Science Board panel in April 2003 unanimously
recommended promptly re-engining the B-52H
fleet. The $3–3.5 billion cost could be privately
financed by engine-makers as a shared-savings
deal or even a “Power-by-the-Hour™ ”service
(continued on next page)
89
Saving Oil
Box 11: Saving oil in existing military platforms (continued)
lease.437 A decision is still pending, but on 25
May 2004, DoD’s proposed $23.5b 100-tanker
acquisition was suspended for further review.
Navy: communities afloat
The Navy has led all Services in institutionalizing energy savings—partly by letting skippers
keep for their own ships’ needs up to 40% of the
fuel dollars they saved.438 Naval energy savings
of 26% during 1985–2002, saving a half-billion
dollars, included such innovations as installing
fuel-flow gauges and performance curves to
optimize cruise speed within mission requirements, and shutting off unneeded engines.
The Navy is on track to save 35% by 2010.439
Just the FY2001 efforts saved 1 Mbbl worth $42
million—enough fuel to support 38,000 steaming
hours, and equivalent to getting free fuel for
19 destroyers.440 But there’s far more still to do.
Visiting several surface and submarine vessels
during the 1990s, this report’s senior author
noticed many Naval design details as inefficient
as those in civilian buildings and equipment,
despite the sixfold higher cost of onboard electricity. Of course, the Navy has unique design
imperatives: ships must go far and fast through
all the world’s climates, project power, protect
crews, and fight through gales and missile
strikes. Being shot at demands serious redun437. Carns 2002.
438. The Incentivized Energy Conservation Program (ENCON)
(www.i-encon.com), introduced in the early 1990s and fully adopted
in FY2000, is described by Pehlivan 2000. Of measured savings,
up to 40% go as cash awards to the ships, 10% for ENCON training
and program administration, and 50% to regional commanders to
improve ships’ readiness. In FY2000, the Atlantic Fleet stopped paying the cash incentives, which ENCON hopes to reinstate soon.
The Pacific Fleet currently pays awards totaling $2M/y to ships that
“underburn” their baseline fuel-consumption target. Minor cash
bonuses also come with awards for which both Fleets compete,
and other forms of recognition encourage competition between
ships and between Fleets.
439. Navy Information Bureau 613 2002.
440. See Encon 2002.
90
441. Lovins et al. 2001.
dancy and special operational methods.
Cramped space often requires small and twisting pipes and ducts, especially when those that
get installed second must snake around those
that got installed first. Inefficient modes of running equipment may be required for prudence
under certain threat conditions or operational
requirements. Nonetheless, the Secretary of the
Navy and the Deputy Chief of Naval Operations
suspected an opportunity. In 1999–2000 they
therefore invited Rocky Mountain Institute (RMI)
to examine “hotel loads” on a typical surface
combatant—the USS Princeton, a billion-dollar
Aegis cruiser then in the top efficiency quartile
of her class, and burning ~$6 million worth of oil a
year, a third to a half of it to generate electricity.
RMI’s engineers found nearly $1 million/y in
retrofittable hotel-load and operational savings
in such mundane devices as pumps, fans,
chillers, and lights and in the mode of operation
of their gas-turbine power generators—uses
that consume nearly one-third of the Navy’s
non-aviation fuel.441 Such a saving per hull
would extrapolate to ~$0.3 billion/y for the
whole Navy. The potential savings RMI found
were several times the 11% previously estimated by the able engineers at Naval Sea Systems
Command (NAVSEA). If fully implemented, RMI’s
recommendations could save an estimated
20–50% of Princeton ’s electricity and hence
10–25% of her fuel (perhaps even 50–75% if
combined with other potential improvements in
power generation and propulsion), while
extending range, stretching replenishment intervals, reducing signatures, and moderating
machinery wear and crew heat stress. And this
doesn’t count a further 8% savings potential
NAVSEA had already found in the propulsion,
power, and combat/command systems that RMI
didn’t examine. An RMI suggestion for both retrofit and new-ship experiments is under review.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Military vehicles: Military efficiency potential
savings, with feasibility shown by 2012, and that by 2000 the Army
already considered this “achievable for combat systems.”442 For example,
today’s armored forces were designed to face Russian T-72s across the
North German Plain, but nowadays their missions demand mobility.
Only one ~70-ton tank fits into the heaviest normal U.S. lift aircraft,443 so
deployment is painfully slow, and when the tank arrives in the Balkans, it
breaks bridges and gets stuck in the mud.444 Army Research, leapfrogging
beyond 20- to 40-ton concepts, has proposed a novel 7–10-ton tank 445 that
uses ~86% less fuel (~4+ mpg), yet is said to be as lethal as current models
and (thanks to active protection systems) no more vulnerable.446 The Army
reckons that such redesign could free up about 20,000 personnel—a whole
division plus their equipment and their own logistical pyramid—that are
needed to deliver fuel to and in theater. This would save $3-plus billion a
year for theater forces, plus more costs upstream.
Yet even this transition may be only a partial interim step. A National
Research Council exploration of initial Army After Next (AAN) concepts
analyzed deployment of an 8,000-soldier division with ~2,000 vehicles,
none exceeding 15 tons. Despite having 30% fewer troops, 48% fewer
vehicles, and 30% less weight than a mid-1990s “light” infantry division—or 53% fewer troops, 75% fewer vehicles, and 88% less weight than
a “heavy” armored division—the hypothetical AAN division, inserted
near-vertically by rotorcraft or tiltrotorcraft, could still need, just to get to
the battle area, fuel weighing three times as much as the entire battle force.447
This suggests the potential logistical value of even more radical lightweighting, possibly to a <1-ton ultralight tactical/scout expeditionary
vehicle analogous to the Revolution composite car (Box 7).448 A little-known
Saving Oil
Army Research has
proposed a novel
7–10-ton tank that
uses ~86% less fuel,
yet is said to be as
lethal as current
models and no more
vulnerable. Such
redesign could save
about 20,000 personnel needed to deliver
fuel to and in theater,
saving upwards of
$3 billion a year.
442. Salyer 2000, slide 23.
443. That is, C-130 or C-17,
which can both use short
runways; in principle, a
C-5 Galaxy can carry two
Abrams tanks, but it needs
long runways and is old and
unreliable. C-130 can lift
only a 20-ton vehicle. C-5
moved 48% of all cargo to Afghanistan and Iraq in Operations Enduring Freedom and Iraqi Freedom, but as all Services strive to get “light, lean, and lethal,”
such heavy lifting, “some time in the next 10–15 years,…could go away entirely as a requirement,” says General Handy, Commander of U.S. Transportation
Command Air Mobility Command. This is because the Army is trying to move toward equipment that can all be lifted by a C-130 or less. However, the Air Force
is modernizing its C-5 fleet because not only lift weight drives its requirement (Tirpak 2004).
444. Newman 2000. Abrams is also too big and heavy to cross the Alps overland (due to bridge and tunnel constraints), so it’s faster to deploy them to the
Mediterranean by sea from the United States than by land from Germany.
445. Using ultralight composite structure and armor, it would be far more capable and protective than the useful family of ~10-ton-class metallic light tanks
such as the British Scorpion, German Wiesel, and the U.S. M8 AGS and M113 Gavin (neither currently fielded by U.S. forces).
446. That doesn’t mean invulnerable. Recently developed anti-armor weapons can reportedly defeat all known and projected forms of armor made of any
materials using currently known forms of atomic and molecular bonds, even at prohibitive (meters) thicknesses. In perhaps a decade, such anti-armor technology will probably diffuse to the cheap, portable, and ubiquitous status of today’s RPGs, which are themselves a significant threat, with a million tank-busting RPG-7 s in use in 40 countries, each selling for under $1,000 (Wilson 2004). In the coevolution of arms and armor—despite great effort and much progress
with active, reactive, smart, electromagnetic, dynamic, biomimetic, and other techniques—we are getting to the point where the best defense is a good
offense, combined with situational awareness, agility, and stealth. This realization is one of the forces driving military transformation’s move away from
heavy armor; nondeployability and logistics cost and risk simply reinforce that conclusion.
447. NAS/NRC/CETS 1999. However, the panel noted potential mitigation, including, over water or very smooth terrain, the possibility of the Russian wing-inground concept, akin to the low-energy flight of large waterfowl just above the surface (NAS/NRC/CETS 1999, p. 68; Granger 2003, pp. 58–59).
448. Such a “HyperVee” platform in military use could get >60 mpg with a diesel hybrid or >90 mpg with a fuel cell, yielding ultralow sustainment, signatures,
and profile. Being fast, small, and agile, it could hide behind little terrain, and its composite structure could resist small-arms fire and shrapnel with no extra
weight, supplementable by light composite appliqué armor. Being thin-skinned, its tactics would presumably include UAV reconnaissance and other tools for
situational awareness. It could be made occupant-liftable and field-refuelable; could carry formidable PGM or recoilless weapons; could be made air-droppable and amphibious; and could be highly deployable, with two personnel able to load ~20 weaponized units into one C-130. A fuel-cell version could also
make ~2.5 gal of pure water per 100 miles, solving the second-biggest sustainment problem—getting drinking water to the warfighters.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
91
Saving Oil
Option 1. Efficient use of oil: Transportation: Military vehicles: Military efficiency potential
When 30 Abrams
tanks were set
against 30 Baja
dunebuggies armed
with precisionguided munitions,
the prompt result was
27 dead tanks (21
completely immobilized) and three dead
dunebuggies.
1982 Army experiment suggests possible tactical value too: when 30
Abrams tanks were set against 30 Baja dunebuggies armed with precisionguided munitions, the prompt result was 27 dead tanks (21 completely
immobilized) and three dead dunebuggies. (In a subsequent experiment,
missile-toting dirtbikers apparently outgunned the tanks even worse.449)
That exercise was done in a desert, not in a forest or city, not under chemical warfare conditions, and reportedly with unimpressive tank tactics,
but it’s still instructive. With different tactics, light and even ultralight
forces, being fast, hard to see, and even harder to hit, may be more effective than familiar heavy forces—especially if they can get to the fight
quickly and keep fighting with little sustainment while the tanks and
their fuel are still en route.450
Such information-enabled breakthroughs are at the core of today’s military transformation strategy. A light, agile Army (and a Marine Corps
unencumbered by heavy armor) could greatly reduce the burdens on the
Navy’s and Air Force’s lift platforms and their fuel use to insert, sustain,
and extract those forces, creating a virtuous circle of oil savings. The
Navy is finding similarly important opportunities for speedier deployment and resupply: a 1998 wargame plausibly equipped the 2021 Navy
with a 500-ton ship that can carry a million pounds for 4,000 nautical
miles, and with ships that can travel at 75 knots.451 U.S. and Russian
designers in 1999 were exploring ships, like Tasmanian wavecutters, that
can carry over 10,000 tons for more than 10,000 nautical miles at more
than 100 knots.452 And light forces could greatly facilitate Sea Basing.
449. An interesting compendium on bicycle warfare
is at LBI, undated.
450. Modern doctrine for
Future Combat Systems is
to avoid encounter (through
situational awareness and
tactics), avoid detection
(signature management),
avoid acquisition (same
plus electronic countermeasures), avoid hit (same
plus active protection),
avoid penetration (lightweight composite armor),
and avoid kill (TACOM 2003).
Collectively, these stages of
protection are intended to
substitute for thick, heavy,
but increasingly vulnerable
passive armor.
451. Hasenauer 1997.
452. NAS/NRC/CETS 1999,
p. 134.
92
A final conceptual example illustrates the scope for major force multipliers that could be more than paid for by their fuel savings and that may
even reduce capital costs up front. The Navy designs ships to carry
weapons systems. RMI’s Princeton survey found that onboard electricity,
made inefficiently in part-loaded gas turbines, costs ~27¢/kWh, so the
present value of saving one watt is nearly $20—not counting the weight
or cost of the fuel and electrical equipment, just their direct capital and
operating cost. But saving a pound of weight on a surface ship typically
saves ~5–10 pounds of total weight because the engines, drives, and fuel
storage systems shrink commensurately. In a ship, such “mass decompounding” is less than in an airplane but far more than in a land vehicle.
So saving a watt must be worth much more than $20 present value. What
is it worth, therefore, to design a watt of power requirement, or a pound,
or a cubic foot, out of a Naval weapons system? Nobody knows, because
the question hasn’t been asked, so weapons systems have clearly not been
so optimized. But organizing naval architects’ “design spiral” around
such whole-system value should yield far lighter, smaller, cheaper, faster,
and better “Hyperships.” This requires changes in the stovepiped design
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Transportation: Military vehicles: Military efficiency potential
culture—so that whole systems are optimized for multiple benefits (not
isolated components for single benefits) purging tradeoffs and diminishing returns, eschewing incrementalism, and rewarding the whole-system
results we want. As with automaking, this isn’t easy. But it’s important to
do before others do it to us.
Recent battlefield experience suggests that the Joint Chiefs’ transformational doctrine emphasizing light, mobile, agile, flexible, easily sustained
forces is vital to modern warfighting. It’s very far from most of the forces
now fielded; heavy-metal tradition dies hard; porkbarrel politics impedes
fundamental reform. But warfighters increasingly see rising risk and cost
in vulnerable and slow logistics,453 compounded by cumbersome procurement that cedes America’s technology edge to adversaries who exploit
Moore’s Law by buying modern gear at Radio Shack. Innovations to turn
these weaknesses into strengths could save prodigious amounts of oil,
pollution, and money. We estimate that comprehensive military fuel efficiency could probably save upwards of ten billion deficit-financed dollars
a year—plausibly several times that (p. 267) if we fully count the scope
for redeploying personnel, avoiding vast pyramids of fuel-support personnel and equipment, and achieving the full force multipliers inherent in
wringing unneeded oil out of the whole DoD asset base.
Whether such innovations also make the world more secure depends on
how well citizens exercise their responsibility to apply military power wisely and create a world where its use becomes less necessary. If we get that
right, we can all be safe and feel safe in ways that work better and cost
less than present arrangements, and fewer of the men and women in the
Armed Forces need be put in harm’s way. Military leadership in saving
oil is a key, for it will help the civilian sector—by example, training, and
technology spinoffs—to make oil less needed worldwide, hence less worth
fighting over. We’ll return on p. 261 to that vital geopolitical opportunity,
which could prove to be DoD’s greatest contribution to its national-security
mission. If our sons and daughters twice went to the Gulf in ~0.5-mile-pergallon tanks and 17-feet-per-gallon-equivalent aircraft carriers because we
didn’t put them in 29-mile-per-gallon light vehicles, that’s a military and a
civilian problem—one that both communities must work together to solve.
Feedstocks and other nonfuel uses of oil
One-eighth of U.S. oil is used to make materials, not burned as fuel.
More than a fourth of that “feedstock” is used to make asphalt for roads,
roofs, and the like. The United States is projected to have 6.34 million
lane-miles of paved highways in 2025, 99.6% of them paved mainly with
asphalt, a bitumen-rich refined product. Our economic, civil engineering,
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
If our sons and
daughters twice went
to the Gulf in
~0.5-mile-per-gallon
tanks and 17-feetper-gallon-equivalent
aircraft carriers
because we didn’t
put them in 29-mileper-gallon light
vehicles, that’s a
military and a civilian
problem—one that
both communities
can help solve.
453. Quartermaster
Professional Bulletin 2002.
Half of 2025 asphalt,
and $8 billion a year
in paving costs, can
be saved by mixing it
with rubber crumbs
from old tires, making
roads’ top layer thinner but longer-lived.
Uncertain but large
savings are also likely in synthetic materials made from oil.
93
Saving Oil
454. In 1998, 270 million
tires were scrapped in
the U.S., with 70% being
recycled or exported
according to EERE 2003a.
Estimates for tire stockpiles
in the U.S. range upwards
from 500 million, or 6 MT of
rubber. Our analysis finds
plenty of available mass of
tires even if they become
much lighter in State of the
Art light vehicles.
455. EIA’s 1998 Manufacturing Energy Consumption
Survey (MECS) data (EIA
1998) imply that the inferred
use of ~423 trillion BTU of
petroleum feedstocks by
the plastics sector was
~48% of the 886 trillion BTU
of total 1998 petroleum
feedstocks other than for
asphalt, road oil, and lubricants (EIA 2003c, Table
1.15). We assume this fraction prevails through 2025,
implying a 2025 plasticsindustry petroleum input of
0.99 Mbbl/d.
Option 1. Efficient use of oil: Feedstocks and other nonfuel uses of oil
and materials-flow analysis finds that a mixture of asphalt with crumb
rubber from old tires454 can, if fully implemented, replace virtually all of
that surfacing asphalt. This asphalt-rubber mixture has long been proven
in service from Arizona to Alberta, and has become cost-competitive since
its patent expired in 1992. If fully implemented by 2025, it can save 60%
of paving asphalt, 51% of total asphalt in all uses, or 0.36 Mbbl/d of
crude oil, at an average net cost of negative $64/bbl (Technical Annex, Ch.
14). That is, asphalt-rubber paving costs less than traditional asphalt
paving because it uses a thinner paving layer that also lasts longer, resulting in lower materials- and handling-costs. This decreases total highway
agency cost per lane-mile by ~15%, and would save agencies ~$7.7 billion/y in 2025—even more if combined with other paving innovations
that make surfaces more pervious, resin-enriched, or light-colored.
Moreover, with our State of the Art technologies, switching to light-colored
pavements can avoid 0.13 percent of U.S. natural-gas use in 2025, because
by reflecting the sun’s heat they reduce air-conditioning loads that are
met on hot afternoons chiefly by inefficient gas-fired combustion turbines
(pp. 113–114). The reduced “urban heat island” effect valuably reduces
photochemical smog formation. By staying cooler, the pavement also lasts
longer and reduces fuel use for vehicular air-conditioners. Asphalt-rubber
pavement cracks less and reduces both noise and skids; major accidents
fall by two-fifths, or on wet days by half—a valuable benefit to which we
assign no economic value.
On the assumption that petrochemical feedstocks will grow in proportion
to U.S. industrial output and be trimmed in proportion to industrial directfuel intensity, EIA projects that petrochemical feedstocks (excluding
asphalt, road oil, and lubricants) will be 9% of 2025 U.S. oil use (including
LPG), or 2.5 Mbbl/d. A large portion of all petrochemical feedstocks go to
manufacture plastics (apparently ~48% in 1998).455 And ~63% of the energy
consumed by plastics is for feedstocks, the remainder to fuel their processing.456 Together, plastics-industry feedstocks and their process fuel now use
~4.5% of total U.S. energy.457 Yet U.S. use of plastics per dollar of GDP
shows signs of saturating just as all other major materials have done.458
Plastics are also an important target for deliberate materials-efficiency
456. The MECS data (EIA
1998) show that in 1998, 394
trillion BTU of energy in all
forms got used in plastics
processing out of the total
1,067 trillion BTU (37%) of
first use energy; the
remaining 63% was thus
feedstocks. We adopt this
split as probably conservative for petroleum feedstocks, whose quantities in
this sector were withheld
by EIA. For the purposes of
these calculations, plastics
are defined as NAICS Code
325211 only—probably an underestimate, as it doesn’t account for the complex value web of polymers and petrochemical feedstocks. However, 1,067 trillion
BTU is 4.48% of the total first use energy from EIA 1998 Table 1.2, identical to the unsourced American Plastics Council (APC) estimate and consistent with
the 4.0% plastics total primary energy use typical of high-income countries as cited Patel & Mutha 2004.
457. American Plastics Council, undated. APC was unable to provide its source for this information, but EIA 1998 shows 4.48% of U.S. primary energy went as
process energy or feedstock (indistinguishably) into NAICS Code 325211, “Plastic Materials and Resins.” (However, the 1997 Census of Manufactures found
that 14% of plastic materials originate as secondary products of other industries (USCB 1997), and the interlinked flows of materials are so complex that
quantification is very difficult.) Plastics’ large energy use doesn’t mean they’re not worthwhile: in some cases, such as thermal insulation or vehicle lightweighting, using plastics can save far more energy than their production consumes.
458. Ausubel 1998, at Exh. 8, citing Wernick et al. 1996.
459. The total amount of plastics in municipal solid waste (MSW)—25.4 million tons—represented 11.1 percent of total MSW produced in the U.S. in 2001
(EPA, undated).
94
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Feedstocks and other nonfuel uses of oil
gains, if only because of rising costs for disposing of trash, which is 11%
plastic.459 U.S. innovators are already emulating Europe’s and Japan’s packaging reductions, dematerialization, product longevity, and closed materials loops, like DuPont’s billion-dollar-a-year business of remanufacturing
used polyester film recovered by reverse logistics. In 2001, the U.S. recycled
5.5% of its discarded plastics; 460 Germany, 57%.461 We consider Germany’s
57% to be a realistic 2025 U.S. target for State of the Art, and half that rate
(29%) for Conventional Wisdom. Net of the 2000 U.S. plastics recycling rate
of 5.5%, this implies respective 2025 petroleum feedstock savings of 0.61
Mbbl/d (25%) and 0.27 Mbbl/d (11%). Since German recycling is driven
by statutory targets, it can’t be rigorously assumed profitable even though
the targets are surpassed. But its economics will improve with U.S. experience, so we think it conservative to estimate plastics recycling’s 2025 net
Cost of Saved Energy at zero.
Saving Oil
Interface, Inc. has
developed carpets
whose total materials
saving, 99.9%, could
profitably displace
more than 6% of total
U.S. petrochemicals.
Many substitutions away from oil are already occurring invisibly: even
the adhesive in 3M’s Scotch™ Brand tape and in many 3M bandages is
now oil-free.462 This potential to save the petrochemicals industry’s products would be revealed as much larger if we included higher-performance
and more productively used polymers. For example, Interface, Inc. has
developed carpets that are one-third lighter but four times more durable—
an 86% materials saving; can be leased as a floor-covering service—another
80% saving because only the worn carpet tiles are replaced; and can then
be remanufactured with no loss of quality. If fully applied to the synthetic-carpet industry, the total materials saving, 99.9%, could profitably
displace ~6.6% of total U.S. petrochemicals—or ~7.3% counting a wildly
popular family of fractal patterns (randomized like leaves on a forest
floor) that saves 89% of carpet manufacturing and fitting waste, or 9%
of total carpet making, while halving installation and maintenance cost.463
(Natural or biomass-derived materials offer further major potential
for displacing the nylon and other carpet fibers now made from oil.)
We conservatively neglect all such further opportunities because their
complex analysis exceeds available data and this study’s resources.
460. By weight, for the most recent year recorded (EPA 2003a). The 5.5% plastics recovery fraction was, by nearly fourfold, the lowest for any major type of
material in trash, although bottle recovery is somewhat better at 21%. The 2000 plastics recovery rate was 5.4% (EPA 2002a). Lacking good massflow data for
polymer recycling, our analysis assumes that any molecular inefficiencies and additions in recycling polymers will roughly offset petroleum process energy
saved by avoided polymer manufacturing from virgin materials.
461. Includes recycling and energy recovery for 2001. APME 2003; Duales System Deutschland AG 2004. German “extended product responsibility,” mandatory deposits, and other policy innovations now spreading to several dozen countries, raised the rate of packaging recycling from 12% in 1992 to 86% in 1997.
During 1991–97, such initiatives raised plastic collection by 19-fold and cut home and small-business packaging use by 17% (Gardner & Sampat 1998).
German automakers are world leaders in recycling plastics, with unexpected benefits: BMW’s Z-1 thermoplastic skin was not only strippable from the metal
chassis in 20 minutes on an “unassembly line,” but also made repairs much easier (Graedel & Allenby 1996).
462. Uchitelle 2004.
463. Approximately 13 million pounds of carpet go to U.S. landfills daily (Interface Sustainability 2001). This is after accounting for the 4% of carpets that are
currently recycled (Interface) and 1% of carpets produced from natural fibers (University of Nebraska, Lincoln 1996). Scaling this value by 2% per year
(Global Information, Inc. 2003) equals 19.7 million lb/d in 2025, ~95% made from petrochemicals. In 2003, the U.S. plastics industry produced 282 million lb/d of
all polymers.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
95
Saving Oil
Option 1. Efficient use of oil: Feedstocks and other nonfuel uses of oil
Full use of advancedcomposite light vehicles in 2025,
if achieved, would
boost polymer demand
by ~69%, equivalent to
four years’ growth.
This temporary investment of oil in recyclable polymer would
save nearly 100 times
that much light-vehicle
fuel each year.
The 25% State of the Art petrochemicals saving is partly offset (temporarily,
until recycling catches up) by our assumption of a light-vehicle shift from
metal to polymer autobodies, which would (assuming full implementation)
consume 6% of EIA’s projected 2025 polymer production rate (equivalent to
four years’ growth).464 This ~0.1 Mbbl/d increase in feedstock requirements
shrinks the State of the Art net feedstock saving to 0.51 Mbbl/d.465 But on
p. 110, we’ll show that most of the remaining petrochemical feedstocks
can be replaced by biomaterials. Today’s forerunners of such carbohydratebased products include McDonald’s potato-starch-and-limestone replacements for polystyrene clamshells and Cargill-Dow’s polylactic-acid-based
textile fibers.
Lubricants, roughly half in industry and half in vehicles, use 1% of U.S.
oil. Most lubricants can be saved by adopting vehicle technologies that
need less or no oil (e.g., ultralight hybrids with roughly threefold smaller
engines and transmissions, if any, would use correspondingly less oil,
and fuel-cell vehicles would use almost none 466); by re-refining used oils;467
and by using synthetic oils468 that last at least twice as long but cost about
the same. Systematically combining these methods should save about
34% of 2000 lubricant petroleum use in Conventional Wisdom (0.08 Mbbl/d)
and 80% in State of the Art (0.20 Mbbl/d).469 Most of the rest should be
displaceable with biolubricants (p. 110 below).
464. We estimate 303 million light vehicles in 2025,
each with a Body-in-Black™
conservatively containing
the same ~116-kg mass of
carbon-fiber thermoplastic
composites and thermoplastic (assumed all made
from oil) as the 2000
Revolution design. Adjusting for
the nominal 2:1 white-fiber:black-fiber ratio for producing the carbon fiber from polyacrylonitrile (ultimately made from propane), and assuming the nonstructural body panels are pure unfilled thermoplastic, implies ~226 kg of polymer input per vehicle body, or 4.6 MT/y for projected 2025 light-vehicle production.
We scale this by 1.08—the ratio of curb weights for the average new State of the Art light vehicle in 2025 to the 2000 Revolution ’s 856.5 kg, noting that EIA’s
2025 “Small SUV” is only 1% lighter than the average new vehicle. This 1.08 scaling factor yields 244 kg of polymer input to make the average 2025 light-vehicle body in this fashion. For EIA’s 20.4 million new units in 2025, that’s 5.0 MT/y, or 10.3% of current U.S. polymer production (48.5 MT in 2003), equivalent to
four years’ growth in the 80% higher polymer production rate EIA adopts for 2025. Assuming that the nonstructural polymer content of the halved-weight but
generally more polymer-intensive State of the Art light vehicles remains roughly comparable to the 115 kg (7.6% by mass) in the typical 2001 car (ORNL 2003,
p. 4-16), this is a quite modest increase in total polymer production. Over time, such a fleet would become substantially self-sustaining in materials because
the extremely durable autobody polymers chosen can be recycled repeatedly, then downcycled. Further refinements, such as adoption of a true monocoque
when Class A molding and reparability have matured, should further lighten the Body-in-Black. We make no corresponding adjustment for the advanced
composites used by trucks and airplanes because their massflows are relatively small and should be part of forecasted feedstock demand.
465. This forecasted 5 MT/y polymer requirement, or 10.4% of 2000 production of 48.2 MT (APC 2001), would be 6.3% of projected 2025 polymer production,
scaled by EIA’s index of industrial output. We round this down to 6.0% as a conservative allowance for recycling of early models and of manufacturing scrap.
This increases 2025 feedstock consumption by 0.1 Mbbl/d.
466. Lovins et al. (1996) show that an ultralight vehicle using an engine hybrid would eliminate 2, or using a fuel cell would eliminate 5–6 (including 22 L/car-y
of motor oil), of the 14 types of fluids now needed to maintain a typical car (whose fluid inventory is ~80 kg), and would require an order of magnitude less
fluid input per year. Such vehicles are also likely to use sealed superefficient bearings requiring no grease. For our technology assumption we simply reduce
car crankcase-oil usage proportional to mpg, i.e. by 27% for CW and 69% for SOA. This value implicitly incorporates all the relevant savings, assuming that
virtually all the lubricant is crankcase oil.
467. Re-refined used oil is measured against the same standards as virgin oil and can therefore be assumed as a direct replacement. In addition, re-refining
uses only 1/3 of the energy required to refine crude oil, saving both process energy and feedstock (Zingale 2002) Currently, only 10% (Buy Recycled Business
Alliance 2000) of collected used oil is re-refined in the U.S. vs. up to 60% in other countries such as Germany (Buy Recycled Business Alliance 2000).
Therefore, accounting for both re-refining substitution and process energy savings, assuming an increase from 10% to 20% for re-refining in the U.S. for CW
and an increase to 60% for SOA, and a used oil collection rate of 58% as in Germany (International Centre for Science and High Technology 2002), we adopt
an overall 10% CW saving and 48% SOA saving of lubricants remaining after technology efficiency improvements (CW = [10%*58%]+[(2/3)*10%*58%] and
SOA=[50%*58%]+[(2/3)*50%*58%]).
468. Current estimates for synthetic market share range from 6% (Synlubes.com 2004) to 7% (Amsoil, undated). CW assumes that 10% of the lubricant market will
be synthetics in 2025 with a 50% longer life than conventional lubricants, thereby eliminating the demand for half again as many bbl-equivalents each year (15%
total). SOA assumes that 20% of the lubricant market will be synthetics in 2025 with double the useful life or 40% total.
469. See next page.
96
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Saving Oil
Option 1. Efficient use of oil
Industrial fuel
For fuel, as distinct from feedstocks, industry uses distillate fuel oil and
liquefied petroleum gas, plus small amounts of low-value residual oil and
coal-like petroleum coke, to produce process heat and steam. We conservatively estimate that about 11% (Conventional Wisdom) or 19% (State of the
Art) of this 2025 petroleum use can be saved, for $3–4/bbl or less than a
three-year simple payback. Our practical experience in heavy process
plants suggests a much larger potential—CEF’s Advanced case resembles
the lowest savings we’ve achieved—but data to substantiate this across
diverse U.S. industries aren’t publicly available.
We therefore rely on the findings of a detailed and peer-reviewed interdisciplinary study by the five National Laboratories with the greatest expertise in energy efficiency.470 This Clean Energy Futures (CEF) study examined
20-year technoeconomic efficiency potentials. The Labs analyzed industrial
fuel and power savings by subsector in a Moderate case assuming modest
changes in political will and improved use of off-the-shelf technology, and
in an Advanced case with significant adoption of more sophisticated technologies, partly driven by a carbon-permit-trading system that equilibrates
at $50/TC (tonne carbon). We apply the percentage fuel savings from these
respective cases to EIA’s 2025 industrial fuel use to obtain Conventional
Wisdom and State of the Art savings of industrial fuel and electricity.471
The resulting actually implementable oil savings in 2025 also assume that
saving industrial electricity will back out oil-fired generation first, then
gas- and coal-fired generation in proportions determined by a simple load
duration curve (LDC) analysis.472 Details are in Technical Annex, Ch. 15.
Buildings
Our analysis of buildings, which use 5.7% of oil in 2025, also relies on the
CEF study. Our State of the Art instantaneous potential savings in 2025—
25% at $3.3/bbl—comes from scaling the percentage savings for the last
year of CEF’s 20-year Advanced scenario (which assumes normal stock
turnover) to 100% implementation, and assuming that these percentage
savings of natural gas in buildings applied equally to directly used distillate
oil and LPG (reasonable because they’re priced above natural gas).
One-ninth of 2025
oil is used to fuel
industrial processes.
Much of this can be
profitably saved by
furnace and boiler
improvements, thermal insulation and
re-use, combined
production of heat
and power, and other
well-known and
lucrative practices.
The 6% of 2025
fuel-oil use slated for
buildings can be
mostly displaced by
similar efficiency
gains; later we’ll
displace the rest with
saved natural gas or
with bottled gas.
469. The percentage savings were based on a simplified bottom-up analysis of potentials for these three lubricant saving strategies.
470. Brown et al. 2001; Interlaboratory Working Group on Energy-Efficient and Clean-Energy Technologies 2000; Koomey et al. 2001.
471. Our industrial fuel savings are achievable potentials, not the instantaneous potentials reported here for other sectors. It is not possible to ascertain the
full technical potential based on CEF’s data and methods. Since CEF’s savings already account for stock turnover and policy implementation, we apply 100%
of CEF’s industrial savings in our Coherent Engagement policy scenario.
472. In the electricity efficiency analysis, we backed out all oil used in electricity generation except that in Alaska and Hawai‘i. Coal and gas savings are
then split 39%:61%, respectively, based on our LDC analysis, using the average U.S. heat rate to calculate coal and natural gas fuel savings per kWh saved.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
97
Saving Oil
Less than 3%
of oil today makes
electricity;
and conversely,
less than 3% of
electricity is made
from oil, nearly all
from otherwise
useless “residual” oil.
But a combination
of efficient use and
alternative supplies
can profitably
displace virtually
all of that oil.
New coal and
nuclear plants are
unsuited to the
“peaker” operation
needed to displace
most oil- and gasfired power plants,
and in any case can’t
compete with three
better ways to do the
same things.
98
Option 1. Efficient use of oil: Buildings
Our analysis then scales the CEF Moderate case’s percentage savings proportionately to obtain the Conventional Wisdom potential—13% at $1.7/bbl.
For both scenarios, we allocate electrical savings’ primary fuel displacements as we do for industry. Details are in Technical Annex, Ch. 16.
Electricity generation
Many commentators suppose that energy is energy, so a source of electricity,
such as coal-fired or nuclear power stations, can substitute for oil. In general
this is not so, and not only because very few vehicles (the users of 70% of
oil) run on electricity. In 2000, only 2.9% of U.S. oil was used to produce
electricity, and within that 2.9%, only 0.3% of oil used was distillate products;
the rest was tarry residual oil (the very bottom of the barrel—a gooey coallike tar that’s hardly good for anything else, though it can be expensively
upgraded to lighter products instead). Conversely, only 2.9% of U.S. electricity was made from oil. EIA projects this to drop by 2025 to 1.7%, and the
fraction of oil used to make electricity to fall to 1.5% (0.5% in the form of
distilled product). The link between oil and electricity is thus tenuous.
The 0.5 Mbbl/d of oil used to produce U.S. electricity in 2000 (mainly in
New England, Alaska, and Hawai‘i) is projected by EIA to decline to 0.36
Mbbl/d by 2025. In some cases, the most convenient near-term substitute
will be natural gas, which can be freed up by using it more efficiently,
both in other power plants and in industrial and building applications
(see p. 111 below). Our analysis displaces oil used in electricity generation
using a combination of more efficient end-use of electricity in buildings
and industry and substitution of other generating technologies and fuels.
(For example, in Hawai‘i, home to 10% of U.S. oil-fired electricity generation in 2002, the cheapest supply-side alternatives are renewables such as
windpower, and diverse renewables can meet more than half the need.)
Details are in Technical Annex, Ch. 17.
Although little oil is used to make electricity, 16% of U.S. electricity in
2000 was made from natural gas, projected to rise to 22% in 2025. As we’ll
see later (p. 113), saving electricity is a potent way to displace natural gas
that can in turn displace oil in buildings and industry. In theory, new coal
or nuclear power plants could do the same thing, but for two fatal flaws:
they’re technically and economically unsuited to the “peaker” operation
(few hours per year) that most gas- and oil-fired generation does, and
they’re manyfold, even tens of times, costlier to build and run than buying electric end-use efficiency. They’re also uncompetitive compared to
windpower and gas- or waste-fired cogeneration (Box 25, p. 258).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Electricity generation
Making electricity at the right scale for the job often makes it cheaper,
thanks to 207 hidden advantages revealed by careful examination of
financial economics, electrical engineering, and other “distributed benefits” that typically increase decentralized generators’ economic value by
about tenfold.473 That’s not counted by traditional comparisons of cost per
kilowatt-hour. But distributed benefits are real, measurable, and starting
to shift market choice. They can greatly expand the scope for profitably
displacing oil- and gas-fired generation by relatively small-scale means
that can be mass-produced and rapidly deployed.
Saving Oil
So far, we have found a technical potential to cut 2025 oil use by more
than half, or almost 15 Mbbl/d (Fig. 28), if State of the Art end-use efficiency technologies were fully applied by then. That’s enough efficiency to
cut 2025 oil imports by two-thirds (Fig. 29). Conventional Wisdom saves
half as much at half the unit cost or a quarter the total cost, but even the
huge State of the Art savings cost an average of only $12/bbl, less than
half the 2025 oil price of $26/bbl.
More than half of total
projected 2025 oil use
could be saved by fully
deploying the efficiency
technologies that all
cost less than the
projected oil price
($26 per barrel in 2000 $).
Saving the average
barrel costs only $12.
Such big savings could
return oil use to pre1970 levels and cut
imports by 70% to pre1973-embargo levels.
Of the light-vehicle
savings, 56% would
be realized if all light
vehicles in 2025 just
got the mpg of equivalent 2005 hybrids.
Even the most expensive major tranche of State of the Art savings—light
trucks at 59¢/gal or $19/bbl—is less than EIA’s $26 oil price. This comparison omits oil-industry capital expenditures, such as incremental refineries and distribution capacity, that the 5 Mbbl/d of light-truck savings
(equivalent to twice today’s Gulf imports) could avoid.
473. Lovins et al. 2002.
Combined efficiency potential
More than half the total savings comes from light vehicles, and one-third
just from light trucks, which account for over half the growth in oil use to
2025 (Fig. 7). Most of that potential is on the market in 2004. If in 2025 every
car were a 2004 Prius (p. 29) and every light truck were a 2005 Escape, RX
400h, or Highlander hybrid SUV (Fig. 5)—all of which will doubtless seem
charmingly antique in 2025—then the 2025 light-vehicle fleet would save
37% more oil than Conventional Wisdom light vehicles do, proving those
vehicles’ feasibility ~21 years early. Such a hypothetical 2025 fleet replacement with the best MY2005 hybrids would also save 4.5 Mbbl/d. That’s
only 56% as much oil as State of the Art light vehicles would save,474 emphasizing the importance of lightweighting, but it’s still 16% of EIA’s forecasted 2025 total oil use, equivalent to twice 2002 net imports from the Persian
Gulf. That’s not bad for uncompromised vehicles on the 2004–05 market.
It’s also worth recalling the major conservatisms behind this assessment
(pp. 37–39). In our view, failure of 2004 advanced-composites technologies to reach commercial maturity by 2025—~15 years later than we
expect—is less likely than their being supplanted by such laboratory
wonders (in 2004) as carbon nanotubes and glassy (amorphous) metals.475
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
474. The calculation is in
Technical Annex, Ch. 5, and
assumes the SUVs get 29
EPA adjusted mpg.
475. Nanotubes, besides
their extraordinary and
diverse thermal and electrical properties, may cost
less than carbon fiber
today yet offer roughly
10–100 times its strength
per pound (R. Smalley, personal communication, 20
Nov. 2003). Glassy metal,
well described in an article
of that title by B. Lemley
(2004), may likewise cost a
few dollars a pound in bulk,
yet be six times as strong
as steel. Further revolutions, such as synthetic spider-silk, are also starting to
emerge from imitating
nature’s 3.8 billion years of
design experience (Benyus
1997). There may even be
biomimetic ways to make
carbon nanotubes.
99
Saving Oil
Option 1. Efficient use of oil: Combined efficiency potential
If every car on the
road in 2025 were a
2004 Prius and every
light truck were a
2005 Escape, RX 400h,
or Highlander hybrid
SUV, then the 2025
light-vehicle fleet
would save 16% of
EIA’s forecasted 2025
total oil use,
equivalent to twice
2002 net imports from
the Persian Gulf.
And it’s hard to imagine another 21 years passing without the advantages
of smart growth (especially of demandating and desubsidizing sprawl)
becoming unbearably obvious, leading to major shifts in land-use policy in
much of the United States. (These advantages include improved quality of
life, crime reduction, increased real-estate value, and tax relief.) Trying to
foresee efficiency potential 21 years ahead is always a shot in the dark, but
such is the pace of progress today, and the likely pressure of events to
come, that we feel we’re more likely to have under- than overstated how
much oil-saving by 2025 will look obviously practical and profitable.
Fig. 28 might at first seem to show only that the costliest increments
of end-use efficiency approach EIA’s 2025 crude-oil price projection of
$26/bbl: more precisely, that feeding $26/bbl crude oil into the oil value
chain would deliver refined products at a short-run marginal price
greater than the cost of displacing those products through end-use efficiency. But efficiency’s advantage is actually much greater, because we
have not yet counted three further elements of oil’s true societal cost,
of which the first two are marked on the vertical axis of Fig. 28.
Figure 28: Supply curve summarizing Conventional Wisdom and State of the Art technologies for efficient use of oil, analyzed on
pp. 43–99 above, hypothetically assuming full implementation in 2025. We assume EIA’s convention of expressing savings volume in Mbbl/d
of product delivered to end users, totaling 28.3 Mbbl/d in 2025. On the y-axis we use a value-chain adjustment for wholesale and retail
products to normalize CSE to $/bbl crude. This method (pp. 40–41) accounts for differences in production and delivery costs of different
products as well as energy content contained. The total State of the Art efficiency potential could save more than half of projected 2025 oil
use at an average cost less than half of EIA’s projected 2025 crude-oil price. Realistic implementation fractions, rates, and methods for efficient end-use will be discussed starting on p.169 below, and additional oil-saving potential from supply substitutions, starting on p. 103.
45
asphalt
EIA 2025 oil price + volatility + externalities
(illustrating a minimal externality estimate
of $10/bbl)
buildings
cars
Cost of Saved Energy (CSE)
(2000 $/bbl crude)
35
average CSE
(Conventional Wisdom)
= $6/barrel
EIA 2025 oil price + volatility
25
commercial
aircraft
average CSE
(State of the Art )
= $12/barrel
EIA 2025 crude oil price
electricity
feedstocks
and lubricants
15
50% of 2025
baseline use
25% of 2025
baseline use
heavy trucks
(class 8)
industrial
5
light trucks
0
5
10
–5
–60
–65
oil saved by full deployment in 2025
(million barrels product/day)
15
marine
medium trucks
military
rail
Source: RMI analysis described on pp. 44–99 above.
100
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 1. Efficient use of oil: Combined efficiency potential
Saving Oil
• Value of oil’s price volatility: As noted in Box 3, p. 16, properly comparing
oil with efficient use—which is financially riskless because its cost is
locked in from the moment the equipment is installed—requires
adjustment for the financial value of oil’s price risk. The market, at
this writing, values that risk at $3.5/bbl.
• Externalized cost: Externalities—security, avoidable-subsidy, health,
environmental, and other costs that are real and quantifiable but not
included in oil’s market price—add up to some debatable but substantial amount. As noted on pp. 21–22, it’s hard to justify an externality
value as low as $10/bbl, and rather easy to justify a value of tens of
dollars per barrel—comparable to or exceeding oil’s market price.
• Avoidable capital cost: The long-run avoidable capital cost just to 2025
looks relatively modest (~$0.5/bbl), but in the longer run, replacing
reserves and replacing or refurbishing major capital items like refineries and pipelines could cost considerably more.
Thus even the costliest kinds of State of the Art end-use efficiency are
cheaper than the market price of oil and much cheaper than the true social
cost of oil. The present value of the societal saving from fully buying efficiency instead of oil in 2025—i.e., the integrated area between the State of
the Art supply curve in Fig. 28 and a horizontal line extending rightwards
at $26/bbl—is thus much larger than the $870 billion implied by valuing
2025 crude oil just at its private internal cost of $26/bbl (20-year net present value in 2025 at ~$70b/y rate of net earnings). The true societal value
could be severalfold higher. This surplus can be interpreted in any mixture of at least three ways:
• Cost conservatism: Even if efficiency cost far more than we claim,
it’d still be a better buy than oil.
• Quantity conservatism: We’ve left so much money on the table that
we’re not yet contemplating buying nearly as much quantity of oil
savings as would be worthwhile.
Saving half of 2025 oil
use is not only half
the cost of buying it,
but also avoids major
societal costs not
included in oil’s
price. This conservatism could accommodate doubled costs
of saving oil, or large
shortcomings in
the policies we’ll
propose for encouraging private purchasing decisions
that better reflect
society’s long-term
best interest.
• Implementation gap: By constructing the supply curve at a societal
discount rate (a conservatively high 5%/y real), rather than at far
higher implicit consumer discount rates (often 50–70-plus %/y)—
that is, by taking a long rather than a short view—Fig. 28 shows how
savings can be attractive to society even though they may look too
costly to consumers. The policy section starting on p. 169 describes
how to turn this discount-rate arbitrage opportunity into business
profits as well as public goods. The vertical “freeboard” in Fig. 28
represents a large safety margin to accommodate imperfections in
that arbitrage.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
101
Saving Oil
Option 1. Efficient use of oil: Combined efficiency potential
Figure 29: U.S. oil consumption and net oil imports 1950–2025
Oil-saving potential if Conventional Wisdom (yellow) or State of the Art (green) efficiency technologies
were hypothetically fully adopted at a linear rate during 2005–25, vs. EIA’s Annual Energy Outlook 2004
Reference Case projection (red). A realistic adoption rate is graphed in Fig. 33 on p. 123.
actual
total petroleum use
projected
EIA projection
25
you are here
20
Conventional
Wisdom
practice
for real
State of the Art
15
net imports
2020
2010
State of the Art
2000
0
1990
Conventional
Wisdom
1980
5
1970
EIA projection
1960
10
1950
petroleum product equivalent
(Mbbl/d)
30
year
Source: EIA 2003; EIA 2004; preceding RMI analysis.
476. In principle there’s at
least one additional major
option—synthetic fuels
made from coal. However,
we’ll suggest that climatesafe ways to use coal to
make hydrogen, chiefly by
pulling it out of water, are
an even higher-value use
and are more likely to prove
competitive in the more
demanding post-efficiency
marketplace.
So far, we’ve only considered how to use oil very efficiently. That’s enough,
if fully deployed by 2025, to return U.S. oil use and imports to pre-1973
levels (Fig. 29). (Starting on p. 178 we’ll consider how quickly this can
actually happen.) But our quiver still contains three arrows—the three
potential supply-side displacements for the remaining oil use.476 These all
become more effective, rapid, affordable, and attractive when efficient use
has shrunk what they must do, as we’ll see next. Of course, their contribution can’t simply be added to that of efficiency, because whichever is
done first (typically efficiency if we choose the best buys first) will leave
less oil usage for subsequent measures to displace. But properly integrated,
the combined potential is potent indeed.
Fully capturing the profitable end-use opportunities described so far
could return 2025 U.S. oil use and oil imports to pre-1973-embargo levels—
before the oil-substituting supply options to be described next.
102
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Substituting for Oil
Option 2. Substituting biofuels and biomaterials
Liquid fuels made from farming and forestry wastes, or perhaps from
energy crops, are normally considered to offer only a small potential at
high cost. For example, classic ethanol production from corn, which now
provides ethanol oxygenate equivalent to 2% of U.S. gasoline,477 could
expand by only about half by 2025 if not subsidized. Modern production
plants of this type, especially if highly integrated,478 yield net energy,
but need favorable resale prices for their byproducts (mainly distiller’s
dried grains and, in other countries, electricity) to compete with gasoline.
And gasoline, as noted on pp. 20–21, is already rather heavily subsidized.
However, that widely held perspective, reflected in our Conventional
Wisdom biofuels portfolio (Fig. 30a), is outdated. When the National
Academies’ National Research Council found in a 1999 study that biofuels
could profitably provide 1.6 Mbbl/d by 2020,479 new methods of converting cellulose- and lignin-rich (woody) materials into liquid fuels, e.g.
using genetically engineered bacteria and enzymes, were just emerging.
Five years later, even newer State of the Art technologies now permit biofuels by 2025 to provide 4.3 Mbbl/d of crude-oil equivalent at under
$35/bbl ($0.75/gal gasoline-equivalent). Of that amount, 3.7 Mbbl/d is
competitive on the short-run margin with EIA’s projected $26/bbl oil
(Fig. 30b), or more to the extent one counts oil’s subsidies and externalities.
Nearly half of the
remaining oil use
can be displaced by
substituting cheaper
supplies.
Option 2.
New technologies
can produce a robustly competitive 3.7
Mbbl/d of liquid fuels
from biomass, mainly
as cellulosic ethanol.
Such biofuels can
profitably displace a
third of the U.S. oil
use remaining after
efficiency gains.
Foreign precedents
are encouraging.
30a
75
gross Mbbl/d
50
net Mbbl/d
25
0
0.0
1.0
2.0
3.0
4.0
5.0
Conventional Wisdom
biofuel supply (Mbbl/d)
2000 $/bbl at biofuel
refinery gate
2000 $/bbl at biofuel
refinery gate
Figures 30a & 30b: Supply curves for the 2025 full-implementation potential of U.S. biofuels, assuming that none (“gross”) or all (“net”)
of the energy required to run the conversion process is derived from biofuels rather than from other energy sources. These graphs
account for the energy content and end-use efficiency of the biofuels in today’s gasoline- and diesel-fueled engines: when converting
them into their equivalent, first as those retail fuels and then as crude oil, we count 1 gallon of diesel fuel as equivalent to 1.1 of
biodiesel, and 1 gallon of gasoline as equivalent to 1.23 of ethanol.480
30b
75
50
25
0
0.0
1.0
2.0
3.0
4.0
5.0
State of the Art
biofuel supply (Mbbl/d)
Source: RMI analysis, (see text, pp. 103_110, and Technical Annex, Ch. 18).
477. In 2003, 2.8 billion gallons of ethanol were produced, vs. 137.1 billion gallons of gasoline (EIA 2004d, Table 3.4).
478. E.g., Dakota Value Capture Cooperative’s single-site closed-loop project in Sully County near Pierre, South Dakota, combining a cattle feed mill,
feedlot, anaerobic digester, cogeneration plant, and ethanol plant, and exports ethanol, CO2, wet distillers byproducts, liquid fertilizer, compost, and cattle.
See www.dakotavcc.com.
479. NAS/NRC 1999.
480. The conversion rate of 1.23 gallons of ethanol per gallon of gasoline is calculated as follows: ethanol contains only 67.7% of the heat content of gasoline
(84,600 BTU/gal [Higher Heating Value] divided by 125,000 BTU/gal [also HHV]). However, Wyman et al. (1993, p. 875) maintain that, “a 20% gain in engine
efficiency can be obtained relative to gasoline in a well-designed engine.” Therefore, multiplying 67.7% by 1.2 equals 0.812 gallon of gasoline per gallon of
ethanol or inversely, 1.23 gallon of ethanol per gallon of gasoline.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
103
Substituting for Oil
A third or more of
road transportation
fuels worldwide
could be displaced
by biofuels in the
2050–2100 time frame.
New technologies
roughly double the
yields, greatly reduce
the energy inputs,
and often reduce
the capital costs of
classical cornethanol processes.
The Brazilian
ethanol program
provided nearly
700,000 jobs in 2003,
and cut 1975–2002
oil imports by a
cumulative undiscounted total of $50
billion (2000 US$)—
more than ten times
its total undiscounted
1975–89 real investment, and ~50 times
its cumulative
1978–88 subsidy.
Option 2. Substituting biofuels and biomaterials
The ultimately profitable potential is even larger; NRC found it exceeded 8
Mbbl/d by the end of the century. But just the State of the Art 2025 potential, shown in Fig. 30b, is major—13% of EIA’s projected 2025 oil demand
before (or 18% after) fully applying oil’s end-use efficiency potential.481
Our conclusion, detailed in Technical Annex, Ch. 18, is also consistent with
the finding by Battelle Memorial Institute’s Joint Global Change Research
Institute that 9.5 quadrillion BTU/y of biomass energy482 could be provided
without large impacts on the current agricultural system, yielding a few
percent more biofuel at SOA conversion rates (4.6 Mbbl/d at $36/bbl) than
shown in Fig. 30b. Taking a global view, a 2004 IEA biofuels report estimates that “…a third or more of road transportation fuels worldwide could
be displaced by biofuels in the 2050–2100 time frame.”483 And a study for
DoD of how to relieve U.S. oil dependence, like many others lately, recommended a large-scale initiative in cellulosic biomass.484
Of the State of the Art potential, 99% is from ethanol, largely from lignocellulosic feedstocks. The new technologies often use very efficient enzymes
(many but not all from genetically modified bacteria, and the best about
tenfold cheaper than they were two years ago) for both digesting cellulose and hemicellulose into sugars and then fermenting them. Other paths
include thermal processes demonstrated at pilot-plant scale, such as the
Pearson Gasification process, which produces Fischer-Tropsch ethanol
from synthesis gas. (The F-T process connects small hydrocarbon molecules into long chains, produces a zero-sulfur and zero-aromatics synthetic diesel fuel completely compatible with existing infrastructure, and can
be applied to syngas made from any hydrocarbon or carbohydrate.)
Collectively, such innovations roughly double the yields, greatly reduce
the energy inputs, and often reduce the capital costs of classical cornethanol processes. They also offer greater scope for coproducing valuable
tailored biomaterials. The other 1% of the State of the Art biofuel potential
is biodiesel, an ester normally made by reacting an alcohol with vegetable
oil; it too is becoming cheaper, and should soon compete in pretax price
when using the cheaper kinds of feedstocks—especially those which, like
used cooking oil, are often currently a disposal cost. Comparable bio-oils
usable as diesel fuel can also be produced thermally from a wide range of
feedstocks, as noted below, potentially increasing their fraction and the
total size of the biofuel potential beyond that examined here.
481. I.e., 3.7 Mbbl/d divided by EIA’s projected 2025 oil demand of 28.3 Mbbl/d is 13%, and is 18% when divided by the remaining oil demand of 20.8 Mbbl/d
after State of the Art efficiency savings.
482. Smith et al. 2004. Assuming our State of the Art conversion rate of 180 gal ethanol/dry short ton (dt), this equates to 4.6 Mbbl/d. The intermediate price is
$36/bbl crude oil, but unlike ours, is not converted on the short-run margin from the retail product price.
483. This IEA study predicts a post-2010 price for cellulosic ethanol of $0.19/L or $0.72/gal (IEA 2004a, Table 4.5, p. 78)—higher than our predicted State of the
Art price of $0.61/gal ethanol ($0.75/gal gasoline-equivalent), or slightly lower if Table 4.6 (IEA 2004a, p. 79) is correct in labeling the IEA figures as gasolineequivalents. The IEA price is based on an NREL estimate (as quoted in IEA 2000), that assumed an ethanol conversion rate of 112 gal/ton vs. our SOA conversion rate of 180 gal/ton. Substituting the 180 gal/ton rate into the IEA calculation results in a price of $0.57/gal ethanol, which is actually lower than our predicted SOA price.
484. Petersen, Erickson, & Khan 2003.
104
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 2. Substituting biofuels and biomaterials
Substituting for Oil
Ethanol has replaced about one-fourth of Brazil’s gasoline, already elimiBoth ethanol and esterified
nating oil imports worth ~50 times its startup subsidy.
biodiesel have been proven in
widespread use, ethanol typically
in 10–85% blends with gasoline and biodiesel in 2–100% blends with
diesel oil.485 Brazil’s 29-year-old ethanol program is now the world’s
low-cost producer. Using cheap sugar cane, mainly bagasse (cane-waste)
for process heat and power, and modern equipment, it provides a ~22%
ethanol blend used nationwide, plus 100% hydrous ethanol for four million cars.486 The Brazilian ethanol program provided nearly 700,000 jobs in
2003, and cut 1975–2002 oil imports by a cumulative undiscounted total of
$50 billion (2000 US$)—more than ten times its total undiscounted 1975–89
real investment,487 and ~50 times its cumulative 1978–88 subsidy.488
The Brazilian government provided three important initial drivers: guaranteed purchases by the state-owned oil company Petrobras, low-interest
loans for agro-industrial ethanol firms, and fixed gasoline and ethanol
prices where hydrous ethanol sold for 59% of the government-set gasoline
price at the pump.489 These pump-primers have made ethanol production
competitive yet unsubsidized (partly because each tonne of cane processed
can also yield ~100 kWh of electricity via bagasse cogeneration—a national
total of up to 35 billion kWh/y, ~9% of national consumption).
In recent years, the Brazilian untaxed retail price of hydrous ethanol has
been lower than that of gasoline per gallon. It has even been cheaper than
gasoline—and has matched our 2025 cellulosic ethanol cost—on an energy-equivalent basis for some periods during 2002–04.490 Ethanol has thus
replaced about one-fourth of Brazil’s gasoline, using only 5% of the land in
agricultural production. Brazilian “total flex” cars introduced by VW and
GM in mid-2003 can use any pure or blended fuel from 100% gasoline to
100% ethanol, and are welcomed because they maximize customers’ fuel
choice and flexibility.491 (In contrast, the ~3 million “flex-fuel” vehicles now
on U.S. roads, marketed partly to exploit a loophole in CAFE efficiency
standards but seldom actually fueled with ethanol, can’t go beyond the
“E85” blend of 85% ethanol with 15% gasoline.)
Brazilian
“total flex” cars
introduced by
VW and GM
can use any pure
or blended fuel
from 100% gasoline
to 100% ethanol.
485. However, not all U.S. biodiesel is blendable with petroleum diesel. Moreover, Congress has at times defined biodiesel (chiefly to promote certain subsidies) as involving only certain feedstocks (such as virgin vegetable oils—excluding, e.g., used cooking oils and animal tallow), or being esterified with only
certain alcohols (such as methanol to the exclusion of ethanol and others), or requiring transesterification (thus excluding equivalent fuels that remove longchain fatty acids’ carboxyl group by a thermochemical process instead). Such exclusive definitions may make sense for soybean producers and others seeking favorable treatment for their own option, but make no sense for a country seeking to maximize deployment of and competition between different bio-oils
that are equally functional for displacing diesel fuel.
486. Wyman 2004; Goldemberg et al. 2004. The blend is nominally sold as 22% ethanol (range 20–26%), the rest gasoline, while the neat hydrous ethanol is
95.5% pure ethanol and ~4.5% water.
487. Goldemberg et al. 2004.
488. WBCSD 2004, p. 107.
489. Goldemberg et al. 2004.
490. Table 4.4 on p. 77 of IEA 2004a shows that in fact, in mid-2002 and early 2004, Brazilian bioethanol (at ~$0.72/gal gasoline-equivalent) achieved our 2025
bioethanol price of $0.75/gal gasoline-equivalent. See Goldemberg et al. 2004.
491. IPS 2003. McClellan (2004) estimates a ~70% market share for “total flex” vehicles by 2007 in Brazil.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
105
Substituting for Oil
Option 2. Substituting biofuels and biomaterials
Europe
produced
17 times as much
biodiesel
in 2003 as the
United States did.
With such a mature sugarcane-ethanol industry, Brazil is gearing up for
ethanol exports that could reach 9 million tonnes a year by 2010, over half
of it to Japan (the world’s largest ethanol importer in 2003) and a sixth to
the U.S. The main obstacles are import tariffs designed to protect existing
corn-ethanol industries. The U.S. charges 54¢/gal, raising ethanol’s landed East Coast price from $1.00 to $1.54/gal, and Europe charges 38¢/gal,
but the U.S. tariff wall is leaking. Cargill proposes to dehydrate Brazilian
ethanol in El Salvador for tariff-free export to the U.S. under an exception
in the Central America Free Trade Agreement, despite heavy opposition
from the U.S. corn lobby. Peru is about to open a 25–30,000 bbl-ethanol/d
export facility that would be tariff-free under the Andean Trade Preference Act. Meanwhile, China is exploring major investments in Brazil to
produce both ethanol and castor oil or biodiesel for shipment to China.492
Germany has seen
BP and Shell become
the dominant
distributors of
biofuels, while
independent companies like Greenergy
have taken the lead
in the UK by selling
branded biofuel
products through
supermarkets.
Europe produced 17 times as much biodiesel in 2003 as the United States
did, and the EU is demonstrating that a transition to biodiesel is feasible.
European countries place high taxes on transportation fuels (as high as
74% of the UK’s $5.5/gal price for diesel fuel493), but have been able to
implement partial (UK, France) and even full (Germany, Austria, Italy,
Spain) biodiesel de-taxation.494 This makes biodiesel competitive with
traditional diesel fuel and supports bio-oil feedstock producers as their
agricultural subsidies are phased down. In addition, the European
Commission Directive of 2003 established biofuel targets of 2% energy
content of all transport fuel by 2005 and 5.75% by 2010.495 Unsubsidized
cost-effectiveness will continue to be difficult for biodiesel, however, as
competition for feedstocks increases and as prices for the byproduct glycerin fall with increased supply (unless those lower prices elicit major new
glycerin markets).
The biofuels transition already underway will have significant impacts on
its related industries. Fuel standards will force the development of new
relationships among automakers, engine makers, and fuel suppliers in
order to evaluate biofuels’ impacts on automobile engines and their warranties (such as the Volkswagen/DaimlerChrysler/CHOREN Industries
renewable fuels collaboration). European automakers have already
approved biodiesel blends of up to 5% and are reportedly evaluating
blends up to 30%. Retail fuel distribution will probably remain the same,
but the dominant players in the distribution chain may change. Germany
has seen BP and Shell become the dominant distributors of biofuels, while
independent companies like Greenergy have taken the lead in the UK by
selling branded biofuel products through supermarkets and hypermarkets.
492. Bio-era 2004.
493. Calculated from the
June 2004 supermarket price of £0.795/L or £3.01/U.S. gallon (Automobile Association 2004),
converted at the approximate June 2004 exchange rate of £0.55/$.
494. Automobile Association 2004. The favored feedstocks are rather costly: U.S. biodiesel made from canola (called “rapeseed” in Europe) costs ~$60/bbl
on the short-run margin.
495. European Parliament 2003.
106
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 2. Substituting biofuels and biomaterials
Substituting for Oil
In the U.S., the combination of vehicle efficiency and ethanol output analyzed here suggests that by 2025, the average light vehicle’s fuel will contain at least two-fifths ethanol, rising thereafter—even more if ethanol is
imported. To accommodate regional variations on this average, “flex-fuel”
vehicles accepting at least E85 should therefore become the norm for
all new light vehicles not long after 2010. “Total flex” vehicles like those
now sold in Brazil would further increase the potential to accelerate
ethanol adoption and to manage spot shortage of either gasoline or ethanol. In short, many of the fuel-system, commercial, vehicle-technology,
and production developments that the U.S. would need for a large-scale
biofuel program have already succeeded elsewhere; the main shift
would be using modern U.S. cellulosic ethanol conversion technologies.
The input side:
biomass feedstocks and rural economies
Our analysis of all feedstocks adopts the authoritative Oak Ridge
National Laboratory state-by-state analysis of forest, mill, agriculture,
and urban wood wastes and dedicated energy crops.496 ORNL found an
annual potential to produce and deliver 510 million dry short tons (dt)
of biomass at below $54/dt without competing with existing food and
fiber production or creating soil or water problems. For biodiesel, we rely
on the 2002 Oil Crops Situation and Outlook Yearbook from the U.S.
Department of Agriculture.497
The main energy crops we examine for ethanol production are switchgrass and short-rotation woody crops such as hybrid willow and poplar.
Switchgrass is a fast-growing perennial Midwest and Southeast grass that
curbs soil erosion via extensive ten-foot-deep roots, is drought- and floodtolerant, can be harvested like hay once or twice a year, but has nearly
three times the typical yield of, say, Alabama hay. Such crops could often
be profitably grown, preventing the erosion caused by traditional row
crops, on the 35 million acres of arable land in the Conservation Reserve
Program, which now pays farmers an average of $48/acre-y to grow
resource-conserving, soil-holding, non-crop vegetation, preferably in
polyculture,498 rather than traditional crops.499 Our analysis assumes that
Special energy crops
can provide ample
biofuel feedstocks;
diverse waste
streams may add
more. Sound biofuel
production practices
wouldn’t hamper food
and fiber production
nor cause water
or environmental
problems, and can
actually enhance
soil fertility.
496. Walsh et al. 2000.
497. Ash 2002.
Supplemental sources
included Wiltsee 1998
and Duffield 2003.
498. Conservation Reserve
Program/Conservation
Reserve Enhancement
Program (CRP/CREP) lands
are often degraded or
low-quality, so they especially need to be rebuilt via
diverse prairie-emulating
polycultures to fix nitrogen
and regenerate a biota-based nutrient- and water-management system. This normally takes a few years to get well established, but haying can start after
the first year. The main strategy could be co-cropping or strip-cropping diverse ensembles including such legumes as perennial Illinois bundleflower and
biennial sweet clover, and perhaps encouraging warm-season grasses that can associate with nitrogen-fixing bacteria. Ensembles of relatively few plant
types can achieve much of the synergy of mature prairies. Some Great Plains prairie remnants have been hayed continuously for 75 years with no fertilizers
or supplements, yet no apparent decline in yield. To be sure, such unmanaged prairie meadows have severalfold lower hay yields than highly bred, managed, and fertilized switchgrass fields. The latter, though, are often fertilized with ~120 kg/ha-y of nitrogen, vs. ~150–200 for corn, which takes up the nitrogen
only half as efficiently. The basic design questions include whether native prairies’ extraordinary biodiversity, which has sustained them with no inputs
through millennia of vicissitudes, is “overdesigned” for human biofuel needs, and on a net-energy basis, what compromise between switchgrass monoculture and prairie-like polyculture will optimize the mix of yield, cost, longevity, and resilience: W. Jackson & J. Glover (The Land Institute, Salina KS), personal
communications, 25–26 July 2004.
499. Haying is currently allowed on certain CRP and CREP lands, but only once every three years, under considerable restrictions,
and incurring a 25% payment reduction (USDA 2004b). New rules would be needed to allow regular haying of deep-rooted perennials and to encourage
appropriate perennial polycultures.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
107
Substituting for Oil
Option 2. Substituting biofuels and biomaterials: The input side: biomass feedstocks and rural economies
Biofuels improve
urban air quality,
and can reduce CO2
emissions by 78%
for biodiesel or 68%
for cellulosic ethanol.
Properly grown
feedstocks can
even reverse CO2
emissions by taking
carbon out of the air
and sequestering it
in enriched topsoil
whose improved
tilth can boost
agronomic yields.
fuel crops would be grown where they’re at least as profitable as traditional crops, and that no dedicated energy crops would be available from
the Rocky Mountain and Western Plains regions (which probably do have
some arid-land native-plant fuel crop potential).
500. Bridgwater 2003.
501. Assuming 166 Mt/y
of municipal solid waste,
15 Mt/y of sewage sludge,
and 220 Mt/y of feedlot
manure, converted at a
30% mass yield (B. Appel,
personal communications,
7 and 10 August 2004).
502. Woolsey 2004.
503. Schumacher,
Van Gerpen, & Adams 2004.
504. GM et al. 2001,
Fig. 3.6, pp. 3–13. .
We also consider offal and other animal processing wastes using a recently
industrialized thermal depolymerization process that could produce 0.05
Mbbl/d of bio-oil, from $20/t feedstocks, at prices as low as $13.20/bbl.
One such process, owned by Changing World Technologies, is analogous
to the delayed-coker technology used in oil refining, and has recently been
extended to handle many further forms of waste, including feedlot
manure, municipal solid waste, sewage sludge, used tires, and automotive
shredder fluff. Other such thermochemical technologies are possible.500
Successful commercial application of this technology to such diverse feedstocks (not included in our analysis) could potentially produce up to an
additional million or so barrels per day of biobased oil or two quadrillion
BTU/y of other energy forms, effectively closing the loop on a some major
waste streams.501 Some of the feedstock streams, of course, have competing
uses, and the competitive-market economics have many uncertainties, so
we haven’t included these resources in our supply portfolio. On the other
hand, some waste streams now incur tipping fees or other disposal costs,
making their bio-oil economics potentially favorable. Further Europeanstyle restrictions on refeeding animal wastes to animals would increase the
U.S. waste streams requiring and paying for disposal.502
Since biofuels contain essentially no sulfur, trace metals, or aromatics,
they also improve urban air quality, and can reduce CO 2 emissions by
78% for biodiesel503 or 68% for cellulosic ethanol.504 Properly grown feedstocks can even reverse CO 2 emissions by taking carbon out of the air and
sequestering it in enriched topsoil whose improved tilth can boost agronomic yields. Using biofuels as a vehicle for better farm, range, and forest
practices can also help to achieve other goals such as reduced soil erosion
and improved water quality, and can dramatically improve the economies
of rural areas. Our State of the Art analysis takes no credit for saving nearly 181 million tonne/y of carbon emissions via biofuels production, which
under the emerging trading system, if sold for say $10–50/tonne, could
cut the biofuels’ net cost by $1–6/bbl.505 This value alone could increase
pretax farm net income506 by about $26–128 from its 2002 average of just
$43/acre-y. The full value could be even larger, and could be supplemented by other revenues from this strategy, as discussed on pp. 162–165.
505. Assuming a crop yield
of 7 tons/acre and an ethanol conversion rate of 180 gal ethanol/ton, 22 barrels of crude-oil-equivalent per acre could be replaced each year. Using a 0.85
kg/L density for crude and an 84% carbon content results in a 2.56 tonne/acre carbon savings. Therefore, assuming a carbon credit price of $10–50/tonne
carbon yields $1.16–5.82/bbl-equivalent, or ~$26–128/acre-y.
506. Moreover, Lal et al. (1998), summarized by the same authors in Lal et al. 1999, suggest that the water- and nutrient-holding capacity of increased soil
organic content could be worth far more. See also Lal 2003, summarizing scientific literature in which he and his land-grant-university and USDA colleagues
find that restorative and recommended practices could sequester in soil enough carbon to offset a fifth of current U.S. carbon emissions (nearly one-third
including aboveground forest biomass), and could continue to do so for the next 50 years.
108
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 2. Substituting biofuels and biomaterials: The input side: biomass feedstocks and rural economies
Substituting for Oil
Saving nearly 181 million tonne/y of carbon emissions
Even without counting carbon and
via biofuels production could increase pretax farm net income
ecological benefits, biofuels’
by about $26–128/acre-y
domestic and largely rural producfrom
its 2002 average of just
tion could boost not just the
$43/acre-y.
national economy, but especially rural areas’ economy, culture,
and communities. An analysis of a proposal for 5-billion-gallon-a-year
biofuel production by 2012 (1.8 times 2003 ethanol production) found
that cumulatively through 2012, it could save a total of 1.6 billion barrels
of oil,507 cut the trade deficit by $34 billion, generate $5 billion of new
investment and 214,000 new jobs, boost farm income by $39 billion, and
save $11 billion of farm subsidies.508 Similar benefits are becoming apparent in Europe, whose 430 million gallons of biodiesel production in 2003
(vs. 25 in the U.S.) won support from both oil and auto companies, as
well as from farmers and those wishing to reduce farmers’ costly surpluses and subsidies to conform to the 1 August 2004 world trade agreement.
Biofuel development must not exacerbate, but if soundly pursued
could help ameliorate, two problems common in farming, ranching, and
forestry.509 The first is unsound practices that deplete topsoil, biodiversity
(especially in soil microbiota), groundwater, and rural biotic cultures.
The second, due largely to distorting subsidy patterns and to lax antitrust
enforcement against giant grain dealers and packing houses, is unhealthy
market concentration and near-monopsony: by 2002, 62% of agricultural
goods came from just 3% of farms, and most farmers’ margins continued
to head toward zero.510 Driven by a combination of customer food-safety
concerns, soil erosion, litigation, primary-producer economic desperation,
market innovations, and large-scale monoculture’s increasing risks,
a quiet but pervasive grassroots shift has begun. This shift, away from
monocultural cropland, rangeland, and woods is beginning in land-grant
universities, extension offices, and farms, ranches, and forest operations
around the country. The size of this shift is still small, but its economic
logic, and the ecological logic that ultimately drives the economics, is
compelling. It is even beginning, though initially on a more limited scale
than authorized and intended by law, to be rewarded by the 2002 Farm
Bill’s popular Conservation Security Program.511
Treating soil like dirt
is proving less profitable and durable than treating it as a biotic community—
an extraordinarily valuable form of natural capital to be productively used and reinvested in.
507. That saved 1.6 billion barrels of oil is half the nominal mean reserves of the Arctic National Wildlife Refuge,
which couldn’t produce anything by 2012 even if it were approved and economic today.
508. Urbanchuck & Kapell 2002; Renewable Fuels Association 2004.
509. Hawken, Lovins, & Lovins 1999, Chs. 9–11.
510. USDA 2004a.
511. Like commodity price-support programs but unlike any other soil conservation program, the CSP is open-ended in
funding and enrollment, but has so far been artificially constrained by rulemaking that is still in progress and is tracked at
www.mnproject.org/csp/index.htm.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
109
Substituting for Oil
Option 2. Substituting biofuels and biomaterials: The input side: biomass feedstocks and rural economies
Industry standards
of practice and certification procedures
should be designed
into international
biofuels trade from
the start.
In general, treating soil like dirt is proving less profitable and durable
than treating it as a biotic community—an extraordinarily valuable form
of natural capital to be productively used and reinvested in. Highly integrated “Natural Systems Agriculture” is proving that letting free ecosystem services sponsor fertility and crop protection can match or beat the
yields, margins, and risks of practices based on chemical or genomic artifice. Biofuels won’t automatically solve any of the basic problems of contemporary agriculture, any more than farming and forestry reforms will
spontaneously triumph over deeply entrenched practices and concentration trends that are both reinforced by and reinforcing current public policies. But at a minimum, biofuel efforts can support parallel efforts to
achieve these outcomes, if designed around recommended practices.
Biofuel productivity, at home and abroad, could also greatly benefit from
the new nature-mimicking methods’ inherent and increasing advantages.
Industry standards of practice and certification procedures, analogous to
those now spreading through the world’s timber industry, should be
designed into international biofuels trade from the start. This is in
exporters’ interest because it will enable them to be paid for shifting carbon from air to soil, rather than being charged for doing the opposite
(mining soil carbon by unsustainable farming practices).
In 1999, the National
Research Council
predicted biomaterials could
ultimately displace
over 90%
of petrochemical
feedstocks.
Vigorous industrial
activity to exploit
today’s even better
techniques
suggests the
first 1 Mbbl/d is
realistic by 2025.
512–514. Paster, Pellegrino,
& Carole 2003.
515. We adopt a linearly
extrapolated value of 27%
(of 1994 usage) in 2025 from
NAS/NRC 1999. Of the 2.159
Mbbl/d petroleum used as
feedstocks other than for
asphalt and lubricants in
1994 (EIA 2003c, Table 5.11),
27% would be 0.583 Mbbl/d.
Subtracting 0.23 Mbbl/d of
savings due to plastics
recycling (pp. 95–96 above)
from the original EIA feedstock demand of 2.075
Mbbl/d in 2025 leaves net
feedstock demand in 2025
of 1.845 Mbbl/d, of which
0.583 Mbbl/d is a 32% substitution.
Biomaterials
Besides replacing transportation fuels with biobased fuels, there is an
opportunity to replace petroleum-derived materials and feedstocks with
bioproducts—industrial and consumer goods derived fully or partly
from biomass feedstocks (12.3b lb in 2001).512 Organic chemicals, including
plastics, solvents, and alcohols (175b lb in 2001), represent the largest
and most direct market for bioproducts based on the similar basic composition (chiefly carbon and hydrogen).513 Lubricants and greases (20b lb in
2001) are another sizeable market in which bioproducts are beginning
to compete.514
After petrochemical feedstock savings of 0.27–0.51 Mbbl/d through
plastics recycling (pp. 94–95), biomaterials offer a further 0.6 Mbbl/d515
crude-oil displacement from petrochemical feedstocks in our Conventional
Wisdom case and 1.2 Mbbl/d516 in State of the Art. In addition, the considerable variety of biolubricants now emerging makes it reasonable to
target 2025 biomaterials lubricant substitution savings of 0.04 Mbbl/d
for Conventional Wisdom and 0.11 Mbbl/d for State of the Art.517 Details are
in Technical Annex, Ch. 18.
516. For the 2025 SOA biomaterials substitution potential we adopt a conservative 1.2 Mbbl/d—the unused portion of the available biomass after assuming
conversion to biofuels for prices <$26/bbl—for a saving of 55% of the forecasted post-efficiency 2.0 Mbbl/d. The 1.2 Mbbl/d value assumes the same
energy-conversion efficiency as the biofuels processes for the biomaterials processes, i.e., the remaining biomass feedstocks would yield enough biomaterials to displace ~1.2 Mbbl/d of crude-oil petrochemicals demand.
517. The Conventional Wisdom 19% represents a 25% conversion of the 75% of lubricants that are base oil to biobased oils. The 56% State of the Art value
represents a 75% conversion of the 75% base oil. These values were then multiplied by 203,700 bbl/d (EIA 2003c, Petroleum Products Supplied by Type,
Table 5.11, showing 166,000 Mbbl/d; scaled out to 2025 by the forecasted growth rate of 1.2% from EIA 2004, Table 11).
110
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 2. Substituting biofuels and biomaterials: Biomaterials
Substituting for Oil
DuPont’s goal of ~20% biofeedstocks for its giant chemical business
The price of bioproducts remains
by 2010 is a harbinger of the next materials revolution—
generally high compared to those
the shift from hydrocarbons to carbohydrates.
of conventional products, but the
The chemical giants are placing bets consistent with the National
cost gap is closing. Biorefineries
Research Council’s 1999 prediction that biomaterials could ultimately meet
represent a key price reduction
over 90% of the nation’s needs for carbon-based industrial feedstocks.
route for bioproducts, operating
similarly to petroleum refineries by
taking in multiple types of biomass feedstocks and converting them
through a complex processing strategy to a variety of coproducts including biofuels, biomaterials, power, chemicals, and heat. Production on a
commercial scale will also drive down costs as demonstrated by industry
leaders already producing on this scale, including Cargill Dow’s NatureWorks, Metabolix’s PHA polymer production, and DuPont’s 3GT™ polymer platform. The market for bioproducts continues to grow, both by
pure market forces and through such policy mechanisms as mandated
federal purchases of biobased products (notably by DoD), accelerating
the rate at which biomaterials become cost-competitive. DuPont’s goal of
~20% biofeedstocks for its giant chemical business by 2010 is a harbinger
of the next materials revolution—the shift from hydrocarbons to carbohydrates. The chemical giants are placing bets consistent with the National
Research Council’s 1999 prediction that biomaterials could ultimately meet
over 90% of the nation’s needs for carbon-based industrial feedstocks.518
Option 3. Substituting saved natural gas
Overview
Today’s natural gas shortages can be turned into surpluses by using both
electricity and natural gas more efficiently. Our initial estimate is that 12
TCF/y (trillion cubic feet per year) of natural gas, equivalent to 12.7 q/y
(quadrillion or 1015 BTU/y), can be saved in 2025 at a cost of $0.88/MBTU.
Obviously that’s far below the mid-2004 futures market price for natural
gas ($5/MCF delivered in December 2007 519). This additional gas efficiency
potential has been overlooked by policymakers eager to expand domestic gas drilling and liquefied natural gas (LNG) imports. That omission
suggests a risk that costly new LNG terminals, if financed and sited, could
be ambushed by demand-side competition.
The quickest and cheapest way to save large amounts of natural gas
is to save electricity: improving total U.S. electric efficiency by 5% would
lower total U.S. gas demand by 9%—enough to return gas prices to
$3–4/MBTU for years to come. Improved efficiency should start with
the peak loads first, because nearly all onpeak electricity is generated in
extremely inefficient gas-fired simple-cycle combustion turbines (see
discussion below); well-proven techniques could achieve this in just a
few years. Some of the saved gas can be substituted for oil in targeted
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 3.
Today’s natural-gas
shortages can be
turned into surpluses
by efficiently using
both electricity and
directly used gas.
By 2025, natural-gas
equivalent to half of
today’s usage could
be saved at a small
fraction of today’s gas
price. Some of the
saved gas can then
be substituted for oil
in targeted uses,
but substitution via
hydrogen (an option
discussed later)
could save even more
oil and money.
518. NAS/NRC 1999.
519. Wall St. J. 2004a.
111
Substituting for Oil
Option 3. Substituting saved natural gas: Overview
furnaces, boilers, and high-dutycycle vehicles like buses, but most
can yield the greatest value if
used to make hydrogen (p. 227).
Reducing natural gas demand will tame the volatile gas markets, lower
gas prices, and potentially cut gas bills by $40 billion per year and electric
bills by an additional $15 billion per year.520 Electricity savings, detailed
below, will also reduce the likelihood and severity of blackouts by reversing the trend toward larger, more heavily loaded transmission lines,521 by
providing an instantly usable “fire extinguisher” to correct local power
imbalances before they cascade across whole regions,522 and by penalizing
artificial scarcity.523 In summary, the good news is that by adopting these
measures, the U.S. can regain more than adequate North American natural
gas supplies.
Reducing natural gas demand will tame the volatile gas markets,
lower gas prices, and potentially cut gas bills by $40 billion per year
and electric bills by an additional $15 billion per year.
520. RMI’s $40b/y savings
estimate is based on the
nominal annual value of
bringing gas prices down
to $3/MBTU. McKinsey
consultants (Colledge et al.
2002) estimated the $15b/y
power savings from demand
response, but their analysis
only looked at energy costs,
not avoided distribution
capital, so the power-cost
savings are probably twice
as high, especially if the
demand-response investments are even modestly
targeted.
521. Lovins, Datta, &
Swisher 2003.
Recent calls for
LNG imports and
other costly
expansions of U.S.
natural gas supplies
overlook larger,
cheaper, and faster
opportunities to
use gas far more
efficiently.
The degree of future gas substitution will ultimately depend on its price
relative to the price of residual and distillate fuel oil, and well as on any
equipment-related costs of switching. While we can reasonably surmise
that both gas and oil prices would decline as U.S. demand drops (unless
overtaken by some combination of unexpectedly high oil demand elsewhere, supply disruption, and depletion), we do not believe accurate
predictions are available for how these fuels will be competitively priced
against each other. Nevertheless, we are using EIA’s forecasted future
U.S. fuel prices to be consistent with the overall methodology in this
report. We next summarize the findings of our analysis of potential U.S.
gas savings and substitutions; details are in Technical Annex, Ch. 19.
Saving natural gas
United States natural gas markets are currently deregulated, with wellhead prices set by markets. Customers have open access to the natural
gas transportation system, at tariffs defined by regulation—generally
federal for interstate transactions and state for intrastate transportation
and retail distribution. Gas-fired power plants have dominated new
power generation because they offer high efficiency and lower emissions.
The resulting growth in demand for natural gas,524 combined with disappointing recent North American supply and discovery,525 has eliminated
the supply overcapacity “bubble.” The tight supply/demand balance,
combined with gas seasonality, has led to annualized gas price volatility
522. Lovins, Datta, &
Swisher 2003a, based just on
utility-dispatchable load reductions, such as brief interruptions of electric water-heater and air-conditioner loads that are imperceptible to the customer.
Such radio-dispatched demand adjustments are successfully used in diverse utility systems worldwide, in locales ranging from Europe to New Zealand.
523. Analysis by one of the authors (EKD) shows that if California had installed additional load management equivalent to 1% of its peak load, then in 2000–01,
when some suppliers were withholding power supplies to raise prices, shrewd investors could simply have shorted the power market (bet on lower prices),
activated their load management, dropped prices, averted shortages, and taken more than $1 billion from the miscreants. This is far cheaper insurance than
building new capacity—let alone asking the same firms to do so, thereby increasing their already excessive market power.
524. EIA 2004, p. 47, reports 1995–2002 growth of 43% in gas-fired generating capacity and 31% in gas use. New combined-cycle plants are far more efficient
than older simple-cycle or condensing plants, and most of the recently added ~200 GW of combined-cycle plants outran demand, reducing their capacity
utilization (EIA 2004, p. 48).
525. EIA 2004, pp. 33–45.
112
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 3. Substituting saved natural gas: Saving natural gas
of 60% in recent years, vs. 40% for oil.526 Gas prices have risen from their
previously normal range of $3–4/MBTU to $5–6/MBTU,527 where they
could remain in the absence of either higher supply or lower demand.
These high gas prices are a significant motivator for improving end-use
efficiency, and helped to make U.S. gas demand 6% lower in 2003 than it
was at its peak in 2000.528 Some of that reduction is permanent “demand
destruction” from gas-intensive industries’ seeking cheaper fuel by moving overseas.
Gas consumption is projected by EIA to rise precipitously from 23 TCF
in 2004 to 31 TCF in 2025.529 Since domestic gas production is expected
to be no more than 24 TCF, gas imports could rise to 7 TCF—hence the
recent clamor for LNG facilities, Arctic pipelines, and drilling in lower-48
wilderness areas.530 Is there a better way? What if we take seriously the
repeatedly proven economic reality of competitive market equilibration
between supply and demand, and ensure that end-use efficiency can
actually be bought if it costs less than supply?
There are significant opportunities to save U.S. natural gas
in three primary end-uses:
• Electric power generation
Substituting for Oil
The quickest and
cheapest way to save
large amounts of
natural gas is to save
electricity: improving
total U.S. electric
efficiency by 5%
would lower total
U.S. gas demand
by 9%—enough to
return gas prices to
$3–4/MBTU for years
to come.
526. NPC 2003, pp. 285, 289.
527. EIA 2004.
528. EIA 2004.
529. EIA 2004’s baseline
already includes a 34%
decrease in gas intensity
(gas consumption/GDP).
530. NPC 2003.
• Residential/commercial buildings gas use
• Industrial gas use
Electric utilities
Eighteen percent of U.S. electricity in 2004 is expected to be gas-fired, consuming 5.7 TCF.531 By 2025, if electricity is used no more efficiently than
EIA projects, natural-gas demand for power generation is forecasted to
rise to 8.3 TCF.532 Since nearly all power at periods of peak demand is gasfired and inefficiently produced, nearly 25% of the gas used for power
generation is used for peak power generation.533 The leverage at the peak
is tremendous. A decrease of just 1% of total 2000 U.S. electricity consumption would have reduced total natural-gas use by 2%. This decrease
in consumption is readily achievable. California’s emergency program to
shave peak loads to relieve expected summer 2001 power shortages
exceeded its goal of contracting for ~1.3% of total peak demand savings
over two years. Its implementation began in the first five weeks, and significant savings were realized in the first nine months.534 Fig. 31 shows
Nearly all peakperiod electricity is
made from natural
gas—so inefficiently
that each percent of
peak-load reduction
saves two percent
of total U.S. gas use.
Proven and profitable
electric-efficiency and
load-management
methods can save
one-fourth of projected
2025 natural-gas
demand at about a
tenth of today’s
gas price.
531. EIA 2004. In 2003, the gas-fired fraction fell to 16%.
532. EIA 2004 predicts 8.6 TCF in 2020, falling back to 8.3 in 2025 because higher gas prices are projected to make coal more competitive.
533. In 2000, gas used for peak power generation was ~1.26 TCF; we estimate this will rise to 2.39 TCF by 2025.
534. Two decades ago, the ten million people served by Southern California Edison Company were cutting its forecasted 10-years-ahead peak demand by
8 1/2%, equivalent to more than 5% of the peak load at the time, every year. This cost the utility about 1–2% as much as new power supplies. Today’s technologies and delivery methods are far better.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
113
Substituting for Oil
Option 3. Substituting saved natural gas: Saving natural gas: Electric utilities
Since nearly all power at periods of peak demand
is gas-fired and inefficiently produced, nearly 25% of the gas used for
power generation is used for peak power generation.
The gas-saving leverage at the peak is tremendous.
Figure 31: Saving electricity to save natural gas
For moderate reductions in U.S. use of electricity including onpeak periods,
each percent of electricity saved also saves nearly two percent of total natural
gas consumption. We apply the insights of this 2000 RMI analysis to 2025 below
as a reasonable approximation. (EIA projects the gas-fired fraction of U.S.
electricity to rise only from 18% in 2000 to 21% in 2025. EIA also projects the
combustion-turbine-and-diesel fraction of installed capacity—the kinds mainly
displaced by peak shaving—to rise from 12% in 2000 to 15% in 2025.535)
The graph is based on RMI’s plant-by-plant economic-dispatch analysis,
unadjusted for the subtleties of regional and subregional power flows and
transmission constraints and for some minor uncertainties related to dual-fuel
oil/gas units. Electricity savings here include peak periods, either because
onpeak savings are specifically encouraged or because efficiency is improved
in uses like industrial motors and commercial lighting that operate onpeak as
well as at other times. The curve’s slope is steeper for electrical savings up to
5% because combustion turbines, rather than more efficient (e.g. combinedcycle) units, are being displaced.
% of U.S. 2000 natural-gas
consumption of 23.37 TCF
5
4
3
2
1
0
0
5
10
reduction in total U.S. electricity usage (%)
Source: RMI analysis from EIA and commercial power-plant databases (see caption).
15
this little-known but important
relationship between total U.S. electricity savings (that include peak
periods) and natural gas savings.
By 2025, EIA projects that the U.S.
will have 175 GW of combustion
turbines and 202 GW of combinedcycle units, consuming 8.3 TCF of
gas.536 To reduce this consumption,
we recommend starting with peak
load programs (programs that economically reduce customer electricity demand during a utility’s
peak generation periods). These
are generally cheaper than the
capacity cost of a new combustion
turbine,537 but unlike the turbine,
consume no gas. Well-managed
utilities across the country are
actively pursuing these programs,
and EPRI (Electric Power Research
Institute) has estimated that
demand-response programs have
the potential to reduce U.S. peak
demand by 45 GW—more than 6%
of projected demand.538 RMI conservatively estimates that peakload management programs539 can
shave enough peak load to save at
least ~2% of total electricity consumption, and thus the first ~4%
of U.S. gas use could be eliminated
for negative cost, due to those savings’ capacity value and markethedging value.539a
535. This capacity allocation would be consistent with a peakier load, but oddly,
EIA projects aggregate system load factor to rise from 50.4% to 52.9%. This doesn’t affect our analysis.
536. EIA 2004.
537. For example, the peak-load reductions were expected to cost an average of ~$320/kW (2000 $) over all ten state-administered programs (Ceniceros,
Sugar, & Tessier 2002). This cost would approximate the national average that utilities reported for their 2000 load-management programs. Subsequent
measurement found the California emergency programs’ actual costs averaged ~$155/kW: Rudman (2003) summarizing CEC 2003. (This is for the first 509 MW
of peak savings evaluated, achieved through 2002, and excludes LED traffic signals, which are meant mainly to save kWh, not peak load.) The programs’
cost was decreased by California’s two decades of experience and built-up delivery infrastructure, but increased by major savings previously achieved and
by the premium paid to procure the new savings very quickly.
538. EPRI 2001.
539. This traditional term includes load-shaving and -shifting both under utility control and by customer choice, the latter often informed by real-time price signals
and facilitated by smart meters. Principles and practice are surveyed in NEDRI 2003.
539a. Our economic assessment used the CEF study’s costs, which are for efficiency, not load-shifting; we understand the distinction, and didn’t doublecount the peak-load savings.
114
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 3. Substituting saved natural gas: Saving natural gas: Electric utilities
Electric end-use efficiency programs assumed in the Conventional Wisdom
and State of the Art scenarios will eliminate 18% and 27% of total 2025
U.S. electricity consumption, respectively.540 Such programs, reflecting
good but not best practice and typically excluding side-benefits (such as
reduced maintenance costs that can largely or wholly offset their investment),541 cost around 2.0¢/kWh and 1.7¢/kWh (2000 $, levelized over
the life of the equipment), respectively. If we apply these programs to the
projected 2025 power generation dispatch curve, we estimate that 5.1–8.1
TCF of natural gas could be displaced.542 Subtracting the value of avoided
generating capacity and deferred grid capacity, the CSE is 0.8¢/kWh
for Conventional Wisdom and 0.6¢/kWh for State of the Art. When we convert this to the cost of saved gas, based on the typical efficiencies of the
various types of power plants, we find that State of the Art electric end-use
efficiency could eliminate 25% of 2025 natural-gas use at an average Cost
of Saved Energy of $0.60/MBTU.
Buildings
As explained on pp. 97–98 and in Technical Annex, Ch. 16, we used the
five National Laboratories’ peer-reviewed Clean Energy Futures (CEF)
study543 to calculate Conventional Wisdom and State of the Art instantaneous-potential gas savings in 2025 of 13% (1.28 TCF/y, 1.32 q/y)
and 25% (2.57 TCF/y, 2.64 q/y), at CSEs of $1.51 and $1.71/MBTU,
respectively. These savings, and those below for industry, are all in gas
used directly for heat.
Substituting for Oil
State of the Art
electric end-use
efficiency could
eliminate 25% of 2025
gas use at an average
Cost of Saved Energy
of $0.60/MBTU.
Conservative
National Lab analyses
show a profitable
potential by 2025
to save non-electricutility gas use
equivalent to
another eighth of
2025 natural-gas
demand.
Industrial fuel
Industry uses natural gas to produce process heat and steam, as well as
for feedstocks, which are treated separately below. Based on the CEF
study (p. 97, above), we conservatively estimate that with Conventional
Wisdom technologies, ~5% of 2025 industrial-fuel natural gas can be
saved, equivalent to 0.63 TCF/y (0.65 q/y), at a CSE of $1.50/MBTU.
The corresponding ~11% State of the Art potential is equivalent to 1.33
TCF/y (1.37 q/y) at a lower CSE—only $1.00/MBTU.544 Details are in
Technical Annex, Ch. 15. Our practical experience as efficiency consultants
in heavy process plants suggests a much larger potential, but data to
substantiate this across diverse U.S. industries aren’t publicly available.
The CEF study assumes that the stock of industrial equipment, like buildings, turns over in 40 years and that all the remaining stock gets retrofit540. This analysis is based on the CEF study: Interlaboratory Working Group 2000.
541. Lovins 1994, main text and note 36. Moreover, gains in labor productivity or retail sales are often worth an order of magnitude more than the saved energy
(Romm & Browning 1994; HMG 2003; HMG 2003a; HMG 2003b).
542. Including the displaced oil-fired electricity generation, and assuming nominal national-average grid-loss estimates of 7% offpeak and 14% onpeak.
543. Interlaboratory Working Group 2000. As for oil savings in buildings (pp. 98–98), we simply applied the CEF Moderate and Advanced cases’ percentage
savings to the EIA 2004 Reference Case.
544. Interlaboratory Working Group 2000.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
115
Substituting for Oil
Option 3. Substituting saved natural gas: Saving natural gas: Industrial fuel
ted over 20 years, but it doesn’t show these two effects separately, and
hence understates the savings potentially available from more aggressive
implementation (as is also true of industrial direct oil savings, p. 97, above).
The following table provides a summary of the straightforward potential
gas savings in 2025 from the major sectors we list above.
Table 1: Potential savings of U.S. natural gas via end-use efficiency in electricity, industrial fuel, and buildings, and the corresponding
Cost of Saved Energy. The savings shown are realistically implementable by 2025.
Conventional Wisdom
State of the Art
gas saved
(TCF/y)
CSE
($/MBTU)
gas saved
(TCF/y)
CSE
($/MBTU)
electricity generation
via demand response
5.1
$0.78
8.1
$0.60
industrial-fuel gas
end-use efficiency
0.6
$1.50
1.3
$1.00
residential/commercial buildings’
direct gas end-use efficiency
1.3
$1.51
2.6
$1.71
total
7.0
$0.98
12.0
$0.88
Means of savings by sector
Natural gas savings compound. When less natural gas is delivered
through the nation’s pipeline system, the 3% of gas used as fuel to compress it also decreases more or less proportionally. Conservatively assuming that gas from LNG imports require no pipelining (as imports at least
to Gulf of Mexico terminals would) and are displaced first, pro-rata
compressor savings add a free “bonus” saving of 0.22 TCF/y to the 12
TCF/y of 2025 savings just shown. Moreover, 10–12% of the remaining
compressor fuel gas, or a further ~0.09 TCF/y, can be saved by straightforward efficiency retrofits of existing pipeline compressors at a CSE of
~$0.16/MBTU.545 These two savings increase the total State of the Art 2025
gas-saving potential to 12.3 TCF/y at an average CSE of $0.86/MBTU.
To the extent the saved gas is in fact re-used, however, either to displace
oil directly or to produce hydrogen, rather than being left in the ground,
both of these savings would need appropriate adjustment.
Saving 12 TCF/y of gas, or 39% of EIA’s total projected 2025 gas use,
would reduce gas consumption below U.S. domestic gas production (and
indeed below today’s consumption), avoid the cost of new imports, and
as noted above, cut 2025 gas and electricity bills by probably upwards of
$50 billion a year by reducing both quantities used and prices. The question remains, though: would this lower-price saved gas then actually substitute for oil?
545. B. Willson (Director of Engines and Energy Conservation Laboratory and Associate Professor of Mechanical Engineering, Colorado State University,
Fort Collins), personal communication, 8 April 2004. In addition, much of the compression energy could be profitably recovered—an untapped gigawatt-scale
resource—by reducing gas pressure at the city gate through turboexpanders rather than choke valves. The resulting electricity would probably displace
mainly coal. The obstacles to capturing it appear to be structural and regulatory, not technical or economic.
116
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Substituting for Oil
Option 3. Substituting saved natural gas
Substituting saved gas for oil
Natural gas has served as a substitute for oil for years; in fact, gas
currently used in the end-uses discussed above was at some point substituted for residual fuel oil, distillate fuel oil, or naphtha. Therefore, the
question is not whether this substitution is technically feasible, but
whether it is economical. The decision to substitute gas is based on four
main factors: the difference in fuel prices between gas and the respective
oil product, the retrofit capital costs, environmental regulatory emissions
requirements, and savings due to the gas equipment’s improvement
in end-use efficiency vs. the existing oil equipment. Currently, industry’s
ability to switch from gas to oil is largely limited by environmental
regulations on air emissions, whereas the ability to switch from oil to
gas has been limited by their relative prices at the point of end-use
(the “burner tip”).
Natural gas is no longer the cheap and abundant fuel that it once was.
During the 1980s and early 1990s, gas was priced to be competitive with
residual oil, which in turn is priced roughly 10% below crude oil prices,
so gas remained inexpensive. By 2000, gas demand began to outstrip the
ability of U.S. infrastructure to supply and deliver it, and prices rose to
considerable peaks—up to $10/MBTU.546 During 2001–03, gas has exhibited higher and more volatile prices than oil. Gas futures, too, are currently
trending above oil at $5–6/MBTU,547 with a higher implied volatility than
West Texaco Intermediate crude oil (Technical Annex, Ch. 19).
Regardless of the
unknowable future
relative prices of oil
and natural gas,
saved gas can
profitably displace at
least 1.6 Mbbl/d of
2025 oil, with about
8 TCF/y (equivalent
to one-third of
2004 gas demand)
of saved gas left over.
546. EIA 2004.
547. Wall St. J. 2004a.
While gas can be saved inexpensively (Table 1), it will then be resold not
at its Cost of Saved Energy, but at the market price that reflects the supply
and demand conditions prevailing at the time. Substituting saved gas for
oil is not free, but requires capital costs for the conversion. These investments, absent an environmental requirement, would only be made if the
Table 2: Estimated non-transportation 2025 oil uses, potentially suitable for gas substitution, remaining after full implementation of
State of the Art end-use efficiency and substitution of biofuels, biomaterials, and biolubricants. No transportation uses are considered
except intra-city buses (p. 120).
EIA 2025 projected oil use
(Mbbl/d)
Sector
2025 oil use
after full implementation
of State of the Art
efficiency (Mbbl/d)
2025 oil use after full
implementation of State of
the Art efficiency and
biosubstitution (Mbbl/d)
industrial fuel
2.82
2.28
2.18
petrochemical feedstocks
& lubricants
2.75
2.02
0.79
residential buildings
0.86
0.64
0.64
commercial buildings
0.46
0.34
0.32
intra-city buses
0.05
0.05
0.05
power
0.36
0.04
0.04
total
7.31
5.38
4.03
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
117
Substituting for Oil
Option 3. Substituting saved natural gas: Substituting saved gas for oil
Cogeneration
was a near-universal
U.S. practice in 1910,
but fell into disuse,
making the U.S.
electricity sector’s
fuel productivity
lower in 2004 than
in 1904.
gas commodity price were cheaper or if the gas appliance were significantly more efficient than its oil-fired counterpart. What do these conditions imply about future gas substitution?
To estimate that substitution potential, we assume that the State of the Art
energy efficiency improvements will be implemented for both oil and gas
since they save money for consumers. If their full technical potential is
captured by 2025, and biofuels and biomaterials are then fully substituted, then 4 Mbbl/d of non-transportation oil use will remain in 2025 for
potential gas substitution, as shown in Table 2, p. 121.
12: Replacing one-third of remaining non-transportation oil use
with saved natural gas
Substituting natural gas for industrial fuel oil
Saved natural gas can be substituted for much
of the remaining 2.2 Mbbl/d of 2025 industrial
fuel oil consumption, which is almost entirely
for process heat and (mostly) steam.548 Even
without the lower gas price that might be
expected from saving 12 TCF/y, industrial users
may find switching to natural gas cheaper, either
because industrial-scale gas furnaces are significantly more efficient than their oil counterparts, or because cogeneration (combined heat
and power, or CHP) creates an opportunity to
displace oil-fired steam boilers with heat that
would otherwise be wasted. Our best estimate
is that 56% of industrial oil used as fuel, or
1.2 Mbbl/d, raises steam in non-cogenerating
boilers.549
548. EIA 2004d.
549. Fifty-six percent of residual and distillate fuel oils used in
manufacturing in 1998 were boiler fuel (EIA 1998).
Cogeneration was a near-universal U.S. practice
in 1910, but fell into disuse, making the U.S. electricity sector’s fuel productivity lower in 2004
than in 1904.550 As some U.S. firms and a far larger number abroad demonstrate daily, cogeneration is more energy-efficient than separately
producing power and steam, since the waste
heat of thermal power generation is used to create the steam, replacing boiler fuel. Thermal efficiencies of cogeneration technologies can be
90% or more—higher than a boiler and far higher than a central power station, but cogeneration displaces both. At EIA’s projected energy
prices, a typical 36-MWe industrial cogeneration
unit would have a Cost of Saved Energy around
–$4.52/bbl and an Internal Rate of Return (IRR) of
77%/y (see Technical Annex, Ch. 19). Displacing
1.2 Mbbl/d of industrial oil consumption would
require 43 GW of cogeneration. Is there enough
potential cogeneration available?
550. Cogeneration could save a projected $5 trillion of global capital costs through 2030, $2.8 trillion in fuel cost, and 50% in the incremental power
generation’s CO2 emissions (Casten & Downes 2004; U.S. Combined Heat and Power Association, www.uschpa.org).
551. The CEF analysis was derived for industrial subsectors from Resource Dynamics Corporation’s DIStributed Power Economic Rationale Selection
(DISPERSE) model. The DISPERSE model assumes that U.S. policies are structured to reduce barriers to financing, siting, utility interconnection, discriminatory tariffs, etc., and therefore support cogeneration. For additional information on barrier-busting policies that would promote the growth of
CHP, see Lemar 2001 and Ch. 3 of Lovins et al. 2002.
552. Cogeneration forecast potential based on Lemar 2001.
553. EIA 2001b reports that 72.6% of residences have access to natural gas in their neighborhood, while 57.3% of commercial buildings (67.6% of commercial floorspace) actually use natural gas and 9.7% (9.3% by floorspace) use propane. Presumably more customers may have access to gas than
actually use it, but ~44% of buildings that do not currently use gas or propane could switch. Since more oil is saved through efficiency and biofuels
substitution in the State of the Art scenario, there is less left to be switched to gas (EIA 2001b, EIA 1999).
118
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Box 12: Replacing one-third of remaining non-transportation oil use with saved natural gas (continued)
The CEF study found that 75 GW of new industrial CHP capacity were economically viable in
the Advanced scenario, assuming that U.S.
policies were structured to eliminate barriers
to cogeneration,551 and 40 GW in the Moderate
scenario.552 We thus expect that 1.15 Mbbl/d
with our Conventional Wisdom technologies,
and with State of the Art technologies, at least
56% of the remaining 2025 industrial oil use, or
1.2 Mbbl/d, could be switched to gas through
cogeneration.
Substituting gas in buildings
Since the gas grid does not extend everywhere
in the United States, especially in rural areas,
we estimate that less than half (0.46 Mbbl/d in
Conventional Wisdom or 0.39 Mbbl/d in State of
the Art) of 2025 building oil consumption has the
ability to switch to gas.553 Of this, 0.3 Mbbl/d
and 0.26 Mbbl/d, respectively, is projected to be
residential building demand that would switch
to gas when the existing residential boiler or
furnace must be replaced. We assume that this
switchover will occur because gas furnaces
average one-eighth higher efficiency than their
oil-fired counterparts.554 The Cost of Saved
Energy is –$16 to +$3/bbl for Conventional
Wisdom (average homes) and –$8 to +$18/bbl 555
554. New gas-fired condensing furnaces for residential use have an
Annual Fuel Utilization Efficiency (AFUE) of 94–96%. The equivalent oilfired furnace has an AFUE of 83–86% (ACEEE, undated).
555. There is less oil to displace in homes that have already implemented efficiency measures so their CSE is higher.
556. Cogeneration for commercial buildings is generally on a smaller
scale than for industrial uses. Therefore, the CSE for cogeneration for
commercial buildings assumes a 1-MW electric generating capacity,
vs. 50 MW capacity for industrial applications.
557. NPC 2003.
558. NPC 2003.
559. NPC 2003, II:54.
560. The National Petroleum Council predicts that gas demand for
feedstocks will decrease during 2001–25. However, we do not classify
this decrease as saved gas, because the decrease will be made up for
either by switching to oil or by petrochemical producers’ simply leaving
the U.S.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Substituting for Oil
for State of the Art (high-efficiency homes).
See Technical Annex, Ch. 19, for details.
The remaining oil consumption is in commercial
buildings. Authoritative studies of the relative
merits of gas vs. oil commercial building furnaces and boilers were unavailable at the time
of this writing. However, cogeneration and trigeneration are likely to substitute for oil-fired
heat at almost all locations served by gas, once
the onerous backup charges and interconnection barriers are removed. We make the conservative assumption that commercial buildings’
natural gas access mirrors the residential building sector’s, and estimate that 0.15 Mbbl/d in
Conventional Wisdom and 0.13 Mbbl/d in State
of the Art can be saved through cogeneration,
at a CSE of –$2.77/bbl for both scenarios.556
Substituting for petrochemical feedstocks
Natural gas and natural gas liquids (NGLs,
which EIA counts as part of petroleum consumption, p. 40) are important to the chemical
industry as both fuel and feedstock. Currently,
the U.S chemical industry, making more than
one-fourth of the world’s chemicals, accounts
for 12% of total U.S. gas demand, and 24% of
this feedstock gas consumption is used directly
to make such basic chemicals as ethylene and
ammonia.557 In 1999, when natural gas prices
averaged $2.27/MBTU, the average operating
margin for the chemical industry was 6.8%, but
as gas prices rose to $3.97/MBTU in 2002 (also
in nominal dollars), margins fell to 0.6%.558 The
winter 2000–01 gas price spike idled 50% of
methanol, 40% of ammonia, and 15% of ethylene
capacity.559 Natural gas prices are generally
forecasted by EIA to remain at or above their
high 2004 level, so eroding margins will force
companies to consider either fuel-switching to
oil or moving their operations offshore.560
(continued on next page)
119
Substituting for Oil
Box 12: Replacing one-third of remaining non-transportation oil use with saved natural gas (continued)
Therefore, substituting gas for naphtha feedstocks is unlikely. However, if our gas-saving
recommendations are implemented, natural gas
prices will probably decline substantially due to
decreased demand. (Indeed, gas-intensive
industries can best obtain cheap gas to sustain
their U.S. operations by supporting regional and
national electricity and gas efficiency.) In that
case, companies would probably stay in the
country, thereby sustaining U.S. jobs, but might
still not switch to gas. Refineries will make less
naphtha as they produce fewer refined products, but the naphtha is cheaply available, and
any left over will be made into gasoline.
Shifting natural gas from peaky power-generating loads to steady industrial and petrochemical
loads also has a major hidden advantage: the
industry could better time gas-storage injections, further reducing price volatility. This may
even reduce average prices by reducing buyers’
need to hedge against price volatility and peakperiod deliverability problems in a quite imperfect market. And of course anything that makes
natural gas prices lower and steadier improves
the competitiveness of U.S. industry and reduces
migration offshore, preserving American jobs.
We conservatively omit the following three
further ways to save feedstock natural gas:
• substituting biomaterials for gas just as we
did above (pp. 93–96; Technical Annex, Ch. 18),
to save 0.9 Mbbl/d of oil-derived chemical
561. A Dutch lifecycle assessment found a 31% near-term potential for
improving the 1988 energy efficiency of plastic packaging (Worrell,
Meuleman, & Blok 1995).
562. A 30–50% potential saving available within a decade was found in
the Netherlands (Worrell, Meuleman, & Blok 1995).
563. The reasons for such conversions are often as much to clean up
urban air as to cut fuel and maintenance costs.
564. Powertrain hybridization of intra-urban buses, which are relatively
inefficient because of their slow, stop-and-go service, should save
~30–37% of their fuel, with an increase of 60% in bus fuel economy;
see GM 2004a.
120
feedstocks (e.g., plant-derived polyhydroxyalkanoates have properties similar to petrochemical-based polypropylene’s);
• potential savings in petrochemical feedstocks
(e.g. from plastics recycling561) that would
lower natural gas use, just as it does for oil
(pp. 93–96); and
• using precision farming and organic methods
to reduce the ~0.5 q/y (1998) of natural gas
that goes into nitrogen fertilizer, much of
which isn’t effectively used and simply washes away as water pollution.562
Other uses
The remaining two uses of oil that could use natural gas as a substitute are small (details are in
Technical Annex, Ch. 19). We expect intra-city
buses to switch from diesel to compressed natural gas (CNG) at relatively low cost.563 If CNG
hybrid buses were deployed, 0.07 q/y of natural
gas would displace 0.04 Mbbl/d of diesel fuel at
a CSE of $8–16/bbl.564 Both hybrid and CNG technologies are already commercialized separately—the largest U.S. fleets are respectively in
Seattle and Los Angeles—and we expect their
combined adoption to occur in both Conventional Wisdom and State of the Art. Conversion can
occur rather quickly: Beijing recently converted
its entire bus fleet to CNG and LPG in just three
years. In the electric power sector, the other
main remaining oil use that could in principle be
substituted by gas, end-use efficiency and displacement by renewables eliminates most oilfired electric power generation (p. 98); any minor
potential remaining for gas substitution is neglected here. All the rest is located only in
Hawai‘i and Alaska in areas beyond the gas grid;
as a result, no additional gas-for-oil substitution
is expected, although there is often major potential from efficiency plus small-scale renewables,
both encouraged by high power costs.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Substituting for Oil
Option 3. Substituting saved natural gas: Substituting saved gas for oil
Table 3: Potential 2025 substitutions of saved natural gas (shown in Table 1) for suitable uses of oil after efficiency and biosubstitution
(shown in Table 2). Considerably larger substitutions would be feasible and could be driven by relative burner-tip prices if known or if
influenced by policy (such as carbon trading). The substitutions shown are the minimum expected to be attractive regardless of the
relative prices of oil and gas, with no other interventions. The bus term includes both fuel efficiency (via hybridization) and substitution
of compressed natural gas for diesel fuel.
Conventional Wisdom
State of the Art
oil saved
(Mbbl/d)
gas
substitution
(TCF/y)
CSE
($/bbl)
oil saved
(Mbbl/d)
industrial fuel
1.15
2.89
–$4.52
1.23
3.07 566
residential buildings
0.30
0.50
–$6.44
0.26
0.43
$4.72
commercial buildings
0.14
0.50
$0.22
0.11
0.43
$0.22
intra-city buses
0.04
0.07
$11.86
0.04
0.07
$11.86
total
1.63
3.96
–$4.08
1.64
4.00
–$2.32
Sector
gas
substitution
(TCF/y)
CSE
($/bbl)
–$4.52
Without assuming that saved
Anything that makes natural gas prices lower and steadier improves
natural gas will be resold for
the competitiveness of U.S. industry and reduces migration offshore,
below the burner-tip price of
preserving American jobs.
petroleum products, the State of
the Art methods summarized in
Box 12 can still plausibly displace 1.6 Mbbl/d of oil using just 4.0 TCF/y
of the 12 TCF/y of saved natural gas. The average cost 565 of this displacement is –$2.3/bbl, dominated by the largest term—industrial cogeneration (Table 3).
The potential 2025 natural gas savings in Table 1 and their substitutions
for oil in Table 3 are summarized in Fig. 32 (see next page) as supply
curves.
Gas-for-oil substitution could become considerably greater than shown
in Fig. 32 if driven by a gas price advantage, environmental constraints
(such as ozone restrictions or carbon pricing), or public policy. However,
at the modest one-third substitution level shown here, about 8 TCF/y
of saved natural gas will still be left over in 2025 for a variety of uses.
Two are obvious: combined-cycle power plants (many idled by electric
end-use efficiency) or further co- and trigeneration in buildings and
industry. Both would displace coal-fired electricity, as would become
more likely in a carbon-trading regime. A third use for the leftover gas,
565. CSEs are calculated based on cost of substitution only, except in the case of residential buildings and intra-city buses
where substitution and efficiency gains could not be logically separated.
566. Natural gas used for cogeneration would displace not only oil but also coal and other fuels used to generate electricity.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
121
Substituting for Oil
Option 3. Substituting saved natural gas: Substituting saved gas for oil
for probably the highest profit margins and the greatest savings in fossil
fuel and carbon emissions, would be conversion to hydrogen for use
both in vehicles and in co- or trigenerating fuel cells in factories and
buildings (pp. 227–242).
Figure 32a: Potential savings in U.S. natural gas consumption implementable by 2025 at the marginal
costs shown (from Table 1). Not shown are compressor improvements and flow reductions, petrochemical
feedstock or end-use savings, and displacements of nitrogen fertilizer.
1.8
cost of saved energy
(2000$/Mbtu)
1.6
1.4
average CSE
State of the Art
= $0.88/MBTU
average CSE
Conventional Wisdom
= $0.98/MBTU
1.2
1.0
0.8
0.6
0.4
0.2
0
2
4
6
8
10
12
14
trillion cubic feet/y (TCF/y)
electricity generation
via demand response
residential/commercial
buildings’ direct gas
end-use efficiency
industrial fuel gas
end-use efficiency
Figure 32b: Potential 2025 substitutions of that saved gas for oil still being used (Table 2) after full implementation of State of the Art efficiency and biosubstitution. We show only the substitutions that are cost-effective without assuming that the saved gas will be resold at a burner-tip price lower than that of oil (Table 3).
cost of saved energy (2000$/bbl)
15
10
average CSE State of
the Art = -$2.32/bbl
5
0
0.0
0.2
0.4
0.6
0.8
1.0
1.2
(–5)
petroleum product (Mbbl/d)
1.4
1.6
1.8
average CSE
Conventional
Wisdom
= -$4.08/bbl
(–10)
buses
commercial buildings
residential buildings
industrial fuel
Source: RMI analysis (see pp. 111–121).
122
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Combined
Conventional
Potential
T
o recap: if all State of the Art end-use efficiency recommendations
were implemented by 2025, 52% of EIA’s forecast oil use would be
eliminated. And half of this remaining oil use would in turn be displaced by substituting for oil the available and competitive 2025 biofuels/biomaterials/biolubricants (pp. 103–111), and the clearly advantageous portion of substituting saved natural gas (pp. 111–112). In actuality,
however, the following section, on Implementation, will show that only
55% of that efficiency potential can be implemented by 2025 by the means
described; the other ~45% would remain to be captured soon thereafter.
Given that realistic implementation, one-third of 2025 oil demand would
be displaced by the supply substitutions (pp. 43–102). The idealized-efficiency-only Fig. 29 (p. 102) would then turn into the realistic path shown
in Fig. 33; the two graphs are similar because the supply substitutions by
2025 nearly offset the not-yet-captured efficiency.
Using oil efficiently
and displacing it with
cheaper conventional
substitutes could
meet 80% of forecasted 2025 oil imports.
The rest is less than
what efficiency will
capture soon after
2025. Domestic supply
alternatives could
even displace that
last 20% plus , if
desired (pp. 227–242),
the forecasted
domestic oil output.
Making America
oil-free within a few
decades is thus
both practical and
profitable.
Figure 33: U.S. oil use and oil imports if end-use efficiency, biosubstitution, and saved-natural-gas substitution were realistically
implemented during 2005–25, vs. EIA’s Annual Energy Outlook 2004 Reference Case projection. Leftover saved natural gas isn’t shown.
total petroleum use
petroleum product equivalent
(Mbbl/d)
30
EIA projection
Conventional Wisdom
State of the Art
plus biofuels and biomaterials
plus saved natural gas
substitution
25
20
15
net petroleum imports
10
5
2020
2010
2000
1990
1980
1970
1960
1950
0
EIA projection
Conventional Wisdom
State of the Art
plus biofuels and biomaterials
plus saved natural gas
substitution
year
Source: EIA 2003c; EIA 2004; preceding RMI analysis.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
123
Combined Conventional Potential
We have a recipe for
profitably eliminating
all U.S. oil use,
imported and domestic, over the next few
decades, with
considerable flexibility in both means
and timing. Achieving
this would take only
about as long in the
future as the 1973
Arab oil embargo is
in the past. Within
two generations, a
more prosperous and
secure America could
be oil-free—possibly
without, and certainly
with, a modest part of
the cost-effective
potential from
expanding renewable
energy sources other
than biofuels, notably
windpower converted
to hydrogen.
But how to capture that combined potential? Next we’ll show how innovative business strategies, accelerated by public opinion, can actually capture half of Fig. 29’s efficiency potential and the other half soon thereafter.
But even with 7 Mbbl/d of savings still to be captured after 2025 as vehicle stocks complete their turnover, the 2025 supply-demand balance could
be revolutionized, as charted in Fig. 34. Eighty percent of forecasted 2025
U.S. oil demand—all but 5 Mbbl/d—can be met in that year either by
profitable, actually implementable efficiency and alternative supplies
or by the 7.8 Mbbl/d of domestic petroleum supply that EIA forecasts for
2025.567 Adding the 7 Mbbl/d of further efficiency gains to be captured
soon after 2025 would thus meet the entire forecasted demand without
even needing 2 Mbbl/d of the forecast domestic petroleum output.
And as we’ll see on pp. 227–242, this doesn’t yet count two large further
options—substituting leftover saved natural gas in the form of hydrogen,
or making still more hydrogen from non-biomass renewables.
The 7.8 Mbbl/d of domestic petroleum output shown isn’t actually needed
either. That’s because the 8 TCF/y of leftover saved U.S. natural gas,
plus another 2.5 TCF/y we’ll explain on pp. 238–239, can be converted to
hydrogen (pp. 227–242), which can be used 2–3 times as efficiently as oil.
It can then provide end-use services, such as mobility, considerably
greater than the 7.8 Mbbl/d of oil can do.
Thus we have a recipe for profitably eliminating all U.S. oil use, imported
and domestic, over the next few decades, with considerable flexibility
in both means and timing. Achieving this would take only about as long
in the future as the 1973 Arab oil embargo is in the past. Within two
generations, a more prosperous and secure America could be oil-free—
without even counting any potential from expanding renewable energy
sources other than biofuels.
Having charted this journey beyond oil, how do we begin, conduct,
and complete it? The business and policy opportunities we present next
can take us there, as part of a broader strategy for building a durably
competitive economy, revitalized industries, a vibrant rural sector,
a cleaner environment, and a safer world.
567. Comprising 4.61 Mbbl/d crude oil and lease condensate, 2.47 Mbbl/d natural-gas plant liquids, 0.48 Mbbl/d other refinery inputs (chiefly from natural gas), and 0.24 Mbbl/d for volumetric gain from domestic crude.
124
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Combined Conventional Potential
Figure 34: Petroleum product equivalent supply and demand, 2000 and 2005
Supply-demand integration for 2025, using EIA’s convention of expressing savings volume in Mbbl/d of petroleum product equivalent.
EIA’s forecasted demand in 2025 (with 2000 shown for comparison) could be cut in half (third bar from the left) if State of the Art
end-use efficiency were fully implemented. The following implementation analysis finds that ~55% of that efficiency potential can be
captured by 2025 through Drift plus Coherent Engagement policies, as shown, leaving the other ~45% as the not-yet-captured profitable
efficiency potential shown by the vertical arrow. The 20 Mbbl/d of net demand can then be met as shown by a combination of bioderived oil substitutes, saved and substituted natural gas, and domestic petroleum production, plus 5 Mbbl/d of “remaining supply”
to be derived from any combination of: North American oil imports, biofuel imports, saved natural gas (8 TCF/y of saved gas remains
for substitution either directly or as hydrogen), buying more efficiency or biosubstitutes than shown (since our analysis, especially for
efficiency, stopped short of the forecasted oil price, and far short of the full social value of oil displacement including externalities),
or simply waiting a bit longer to finish implementing the remaining 7 Mbbl/d of State of the Art efficiency, chiefly by completing
the turnover of vehicle stocks.
end-use
efficiency
demand
State of the Art 2025 supply
30
—
25
supply
—
20
asphalt
buildings
15
cars
commercial aircraft
10
electricity
feedstocks and lubes
5
remaining supply
U.S. 2025 oil output (EIA forecast)
substituted saved natural gas
biomaterials and biolubricants
biofuels
2025 savings with Coherent Engagement
SOA savings left to capture
2025 with full SOA end-use efficiency
industrial
EIA 2025
0
heavy trucks
EIA 2000
demand or supply (Mbbl/d)
end-use efficiency
light trucks
marine
medium trucks
military
other
rail
Source: EIA 2003c; EIA 2004; preceding RMI analysis.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
125
Winning the Oil Endgame:
Innovation for Profits, Jobs, and Security
Implementation
Strategic vision
The prize
Imagine a revitalized and globally competitive U.S. motor-vehicle
industry delivering a new generation of highly efficient, safe, incredibly
durable, fun-to-drive vehicles that consumers want. Imagine equally
rugged and efficient heavy trucks that boost truckers’ gross profits by $7.5
billion per year.568 Consider the pervasive economic benefits of an
advanced-materials industrial cluster that makes strong, lightweight
materials cheaply for products from Strykers to bicycles to featherweight
washing machines you can carry up the stairs by yourself. Envision a
secure national fuels infrastructure based largely or wholly on U.S. energy
resources and on vibrant rural communities farming biofuel, plastics,
wind, and carbon. Think of over one million new, high-wage jobs, and the
broad wealth creation from infusing the economy with $133 billion per
year of new disposable income from lower crude-oil costs. Recognize
with pride that with this new economy, the U.S. is nearly achieving international greenhouse gas targets as a free byproduct. Picture increased
energy and national security as oil use heads toward zero, and as the U.S.
regains the leverage of using petrodollars to buy what our society really
needs rather than handing those dollars to oil suppliers to feed an addiction. Imagine a U.S. military focused on its core mission of defense, free
from the distraction of getting and guarding oil for ourselves and the
rest of the world. Imagine that the U.S., able once again to practice its
admired ideals, has regained the moral high ground in foreign policy.
Finally, envision one of the largest and broadest-based tax cuts in U.S.
history from eliminating the implicit tax that oil dependence imposes on
our country by bleeding purchasing power, inflating military and subsidy
costs, and suppressing homegrown energy solutions.
Sound utopian? It is not.
This vision is based on severely practical business solutions to the
“creative destruction” dilemma currently faced by the chief executives
in the U.S. transportation sector, and on the handful of market-oriented
government policies that are needed to help lower the risk of this transition. The business principles are grounded in such classic and prescient
works as Joseph Schumpeter’s writing on the concept of “creative
destruction” (Capitalism, Socialism and Democracy, 1943), Michael Porter’s
Competitive Strategy (1980), and Clayton Christensen’s and Michael
Raynor’s The Innovator’s Solution (2003).
568. Of the ~20b gal/y consumed by trucking, 37% is saved at a net customer gain of ~$1/gal—
the difference in cost between the fuel efficiency technologies and the retail price of diesel fuel.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Since phasing out oil
doesn’t cost money
but saves money,
business will lead it
for profit.
However, supportive
rather than hostile
public policies would
help, and coherent
national-security
policy can be a vital
contributor.
Applying classic
business concepts of
competitive strategy
to the oil problem can
create astonishing
gains in national
prosperity, security,
equity, and environmental quality.
A $180-billion industrial
investment in the next
decade can return
$130+ billion per year
in gross oil savings,
or $70 billion a year in
net savings, by 2025.
Envision the largest
and broadest-based
tax cut in U.S. history
from eliminating
the implicit tax that
oil dependence
imposes on our
country by bleeding
purchasing power,
inflating military
and subsidy costs,
and suppressing
homegrown energy
solutions.
127
Implementation
American business
can continue to be
threatened by
obsolescence and
dwindling market
share, or can grasp
oil displacement’s
unique opportunity
to revitalize key
industries.
If America
intelligently invested
an incremental
$90 billion over the
next two decades
in retooling the
domestic automotive,
trucking, and airplane
industries and
another $90 billion
in domestic energy
infrastructure,
we would have the
capacity to achieve
an oil-free future.
128
Strategic vision: The prize
Our business analysis supports an exciting and astounding conclusion:
intelligently investing an incremental $90 billion569 over the next two decades
in retooling the domestic automotive, trucking, and airplane industries
and another $90 billion in domestic energy infrastructure, could create the
capacity to achieve this oil-free future. Yet despite the high rates of return
on these investments, they entail too much business risk—perceived or
actual—for the private sector to do entirely on its own, quickly enough to
meet national needs.
Vaulting the barriers
Risk-aversion has deep roots in the cultures of very large organizations.
Their enormous sunk costs, both in physical assets and in psychological
habits, create an immune system that stubbornly resists invasion by innovation. This resistance can be shown by rigorous business scrutiny to be
destructive: innovation and competition are the evolutionary pressures
that make the firm stronger. Nevertheless the “not invented here” mentality stubbornly persists. Wrenching change is always difficult and seldom
greeted with enthusiasm. IBM had 77% higher real revenue and 20%
higher real earnings in 2000–03 than it did in 1978–81, the pre-PC age
when it was the king of mechanical typewriters and mainframe computers, but its transition into the microcomputer age was very hard—a neardeath experience. The challenge facing U.S. makers of light vehicles,
heavy trucks, and airplanes is equally daunting. But in a competitive
global marketplace, the alternative to bold leadership is worse. As General
Electric’s former CEO Jack Welch put it, if we don’t control our own destiny, someone else will.
To complement market forces, and to reduce the risk to the weakened U.S.
transportation equipment sector, we need a coherent set of government
policies to support the transition. We need economically efficient policies
that shift companies’ and customers’ choices toward higher-fuel-economy
vehicles of all kinds while expanding their freedom of choice; help manufacturers to retool their factories and retrain their workers; support the
rapid emergence of cost-effective bio-fuels, other renewables, and domestic fuels; upgrade our transportation systems to reduce congestion; align
utilities’ profit motives with their customers’ interests; and eliminate perverse incentives across the domestic-energy value chain. While we can
569. $100 billion is needed for new efficient automotive
and trucking manufacturing capacity, plus R&D for new
platform development. We estimate that ~$30 billion will be
spent anyway in the U.S. to meet projected new car demand,
for a net increase of $70 billion. An additional $20 billion of
incremental investment is needed for airplanes. Therefore,
the total incremental investment is $90 billion—essentially
the same as the ~$91 billion we estimate would be needed
for domestic energy supply infrastructure.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Strategic vision: Vaulting the barriers
Implementation
probably never satisfy those pure libertarians who hold that no government intervention can ever be justified, and indeed we think many of
today’s energy problems spring from ill-advised past interventions, we
will make a reasoned case that some limited, targeted, and carefully
defined changes in federal, state, and local policy (pp. 169–226) are not just
desirable but very important for managing national risks and achieving
national goals.
When considering our suggestions to foster this transition and benefit all
stakeholders, remember that any long-term vision beyond the comfortably
familiar always looks odd at first. In 1900, the U.S. had ~8,000 cars.
Fewer than 8% of the two million miles of rural roads were paved.
Anyone who’d proposed then that half a century hence, the ubiquitous
horse and buggy would be gone, replaced by a wholly unfamiliar infrastructure in which the newfangled oil industry would refine, pipe, and sell
a new product called gasoline, would have been dismissed as a dreamer.
Anyone who’d predicted that a century hence, 170,000 U.S. retail outlets
would be pumping this new fuel into nearly 240 million light vehicles
whizzing along 600,000 miles of highway would have been banished as a
lunatic. Yet we live in that world today because the genius of private
enterprise, building on Henry Ford’s 1908 Model T (which got 2.5 million
cars on the road during 1908–16), was supported by a series of public
policies, from the early decision to have taxpayers finance public roads to
President Eisenhower’s 1956 Interstate Highway System. The changes
proposed here are far less momentous than those. They need virtually no
new infrastructure; they use well-established technologies made by existing
industries; they meet current user requirements even better than today’s
technologies do; and they’re profitable for both producers and consumers.
The issue is how to help them happen smoothly, quickly, and well, so
American industry can vault the obstacles to doing what it does best—
innovation.
In 1900,
the U.S. had ~8,000 cars.
Fewer than 8% of
the two million miles
of rural roads were
paved.
Anyone who’d
predicted a century
hence nearly
240 million light vehicles
whizzing along 600,000
miles of
highway would have
been banished as
a lunatic. Yet we live
in that world today
because of the genius
of private enterprise.
We need economically efficient policies that shift companies’ and customers’ choices toward
higher-fuel-economy vehicles of all kinds while expanding their freedom of choice;
help manufacturers to retool their factories and retrain their workers;
support the rapid emergence of cost-effective bio-fuels, other renewables, and domestic fuels;
upgrade our transportation systems to reduce congestion;
align utilities’ profit motives with their customers’ interests; and
eliminate perverse incentives across the domestic-energy value chain.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
129
Implementation
At many levels and
from many sides,
the Big Three
U.S. automakers
face competition
so unprecedented
and profound that it
could either kill them
or make them
stronger. Their fate
hangs in the balance,
and time is of the
essence.
An Economist cover
story recently
questioned whether
any of the Big Three
would survive the
global hypercompetition of the next
10–20 years.
The endangered automotive sector—
why it’s important to act now
In 1970, U.S. automobile manufacturers were the envy of the world.
The Big Three—Chrysler, Ford, and GM—produced nine out of every ten
cars sold in America. Then the 1970s oil price shocks caught them unawares,
and Japanese manufacturers swooped in, initially offering smaller and
more efficient cars that better met many customers’ needs, then leveraging into other segments. Thirty years later, the U.S. automobile sector is
on the ropes. The Big Three have been reduced to 59% domestic market
share. Their competitive situation is worsening: since 1999, U.S. car manufacturers lost 2% domestic market share each year to Japanese and
European competition, and now barely sell the majority of the cars within
their home market. The last bastion of competitive strength and profits
has been the light-truck sector, where the U.S. still maintains 75% market
share, but this too is now under frontal assault. The U.S. market position
in light trucks is protected by a 25% tariff on imported vehicles, leading
the Japanese and German competition to build their factories on U.S. soil.
These transplants increasingly make all kinds of light vehicles, from
1990s-style truck-based SUVs to the car-based “crossover” vehicles that
outsold them in the first half of 2004 and may portend their decline.570
Reliance on lawyers and lobbyists to try to avert competition and regulation has long restrained American automakers from fully exploiting their
extraordinary engineering prowess.571 And so the nation’s and the world’s
largest industry, providing a tenth of all U.S. private-sector jobs (p. 17)
and creating the machines that provide core mobility for most Americans,
is sufficiently at risk that an Economist cover story recently questioned
whether any of the Big Three would survive the global hypercompetition
of the next 10–20 years.572 On 16 March 2004, its editorialist opined: “Put
bluntly, the short-term outlook for the Big Three is dreadful….If anything,
the long-term outlook is worse.” 573
570. J. White 2004. Schatz & Lundergaard (2004) report that ~40% of people trading in a traditional truck-based SUV in 2004 are shifting to a different vehicle
class or subclass—a quintupling in five years.
571. This incoherence is longstanding: e.g., GM lobbyists killed California’s Zero-Emission Vehicle rule, which had given a head start to the world’s best battery-electric car (GM’s EV-1), just as its marketing was gaining momentum. The few hundred early-adopter lessees, forced to return their beloved cars to be
scrapped, were embittered. Then a 2001 anti-CAFE-standards lobbying blitz strove to convince Americans that efficient cars are unsafe and unaffordable—
unmarketing many of the same OEMs’ key innovations, and infuriating the green market segment, which switched its loyalty to Honda and Toyota. GM and
DaimlerChrysler may sue, with federal support, to overturn California’s proposal to regulate automotive CO2 emissions. Yet to argue that this is a covert form
of efficiency regulation preempted by federal law, i.e. can be complied with solely by raising fuel economy, automakers must deny the feasibility both of biofuels and of their own impressive and heavily advertised hydrogen programs. (This would also reinforce many environmentalists’ suspicions that the White
House’s hydrogen program is a bad-faith stalling tactic, and offend the industry’s technical partners in hydrogen.) Stomping on one’s own cutting-edge
developments can’t advance market acceptance, sales, reputation, morale, or recruitment. Such contradictions between strategic goals, marketing messages, and lobbying or litigating positions are unhelpful, polarizing, and futile. Foreign competitors simply market new products without first needing to stop
unmarketing them. Yet some major environmental groups bear similar self-inflicted wounds: e.g., if the Sierra Club succeeds in forcing EPA to admit authority
to regulate CO2, it could preempt California’s proposed rule.
572. Carson 2003; Maynard 2003.
573. “Buttonwood” 2004.
130
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
The endangered automotive sector—why it’s important to act now
Much has been written about the
causes of the decline of the U.S.
automotive sector and the basis of
competition. To succeed, car manufacturers need lower costs, higher
quality, a breadth of product offerings, a dealer network, and brand
loyalty.574 While U.S. manufacturers have largely closed the cost
gap with their Japanese rivals,
their continued slide in market
share is attributed mainly to persistent gaps in quality and value.
This has translated into a loss of
brand loyalty. In the 1980s, Lee
Iacocca could confidently tell TV
viewers, “If you can find a better
car, buy it.” Later, as competitors
pulled ahead, many buyers continued to follow his advice. In short,
American car companies must
consistently build the cars that
consumers want, with better quality and value than anyone else.575
Yet the industry’s ingrained
practices—from the old strategy
of owning captive rental-car companies to the cultural habit that
employees should normally drive
the company’s products (rather
than driving competing ones to
understand them better)—have
diluted competitive discipline.
And Detroit’s incrementalist technological style, rooted in corporate
culture, has ill served the industry
as Japan’s bolder strategy gave
Toyota and Honda the lead in hybrids.
13: Guilt-free driving: hybrid cars
enter the market
The rapid U.S. growth in hybrid cars illustrates how
quickly products that offer all the key attributes (such as
family-size, peppy, and guilt-free) can claim market
share. Hybrids were introduced in 1999, and, even though
they can cost ~$2,500–$4,000 more than a conventional
car, are expected to gain a market share of 1.5% by 2005,
with a growth rate of a stunning 75% per year. J.D. Power
and Associates believes that the reasons for the rapid
growth are better hybrid technologies and greater customer product line choice, from small compacts to SUVs.
Consumers are seeking to retain their lifestyle with less
environmental impact. It was Toyota’s hybrid, the Prius ,
that swept the industry’s 2004 top awards (p. 29) and
whose 22,000-order backlog has forced some dealers,
facing year-long waitlists, to stop taking orders. By May
2004, with gasoline prices hovering above $2 a gallon, the
55-mpg Prius midsize sedan commanded a $5–8,000 premium in some markets, while U.S. SUVs needed ~$4–5,000
incentives to sell (even Toyota averaged nearly $3,000
incentives across its full product line). Used rental Priuses
are even reselling for more than the new list price.573a
Toyota and Honda have taken the lead in the hybrid market, but U.S. automakers are slowly following and will be
rolling out competitors starting in late summer 2004 with
Ford’s Escape small SUV (p. 31). (EIA forecasts a 5.5%
market share for hybrids in 2025, but assumes that they’ll
average only 33–38 mpg for cars, 27–32 for light trucks—
both worse than any hybrid on the market in 2004.)
573a. Freeman 2004; Carty 2004.
Success in the dynamic light-vehicle market requires relentless focus on
developing attractive product lines that anticipate evolving customer
needs. U.S. preferences for more efficient (and often somewhat smaller)
cars in the 1970s, aerodynamic cars in the 1980s, and sport-utility vehicles
574. Train & Winston (2004) provide an overview of these factors and the relative position of U.S. manufacturers vs. their rivals.
575. Train & Winston (2004) note that while cost was the initial reason for loss of market share to the Japanese, the U.S. car manufacturers consistently
missed product-line opportunities, and had declining quality relative to Japanese and European makers, leading to today’s battle for brand loyalty based on
quality differences.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
131
Implementation
The endangered automotive sector—why it’s important to act now
in the 1990s were all critical trends that defined the basis for growth in
market share. Looking forward, customer preferences for safety, comfort,
entertainment, fuel economy, performance, and environmental friendliness
appear both contradictory and self-evident. Cars are bought as much for
emotional as for economic reasons, and most Americans, it would seem,
want to drive large, sporty vehicles and feel good about it. Manufacturers
of hybrid cars and, imminently, of hybrid light trucks are claiming market
share by recognizing and meeting this customer need (Box 13). Similarly,
many Americans resent the feeling of having to buy an SUV to protect
their families from the heavy, aggressive, incompatible SUVs that increasingly surround them. Makers who first offer family protection without
inefficiency and hostility (pp. 58–59) will win those customers.576
Financial weakness,
incrementalist
culture, and internal
contradictions slow
U.S. automakers’
response to such
obvious forms of
competition as
hybrid-electric
vehicles. But ultralight materials,
software-dominated
vehicle architectures,
and even new
business models
will expand the
competitive threat.
Many obstacles are blocking or slowing U.S. automakers from the fundamental changes they will need in order to prevail in the global marketplace
of the next few decades. Starting on p. 178, we propose a portfolio of market-oriented public policies to break down those obstacles. First we explore
more fully what the challenges are, what causes them, why they’re serious,
and how—like an Aikido master blending with an attacker’s energy—
the industry can turn serious competitive challenges into transformative
opportunities.
Four competitive threats
The first big threat is advanced propulsion technologies. But unlike
Toyota,577 which hopes to expand its global market share from 10% now
to 15% in 2010,578 U.S. automakers’ propulsion strategy isn’t built around
world-class fuel economy: it’s the strategy of an Airbus A380 superjumbo, not a sleek and economical Boeing 7E7 (Box 14). On the whole,
U.S. manufacturers are stubbornly defending their market share in light
trucks by making ever larger and heavier gasoline-engine SUVs, backed
by ever larger financial incentives such as rebates (around $5,000 in mid2004) and interest-free loans.
In contrast, Japanese automakers, especially Toyota, have the financial
strength to bet on both fuel cells and hybrid powertrains, as well as to
invest heavily in both economy and luxury models simultaneously.
Indeed, they’re bringing hybrids’ peppy performance to luxury models,
and already their hybrid powertrains are cheap enough that they’re about
to appear in two of the world’s most popular sedans—the Honda Accord
576. The sense of protection comes from a combination of size, stiffness, strength, and solid handling (e.g., crosswind stability and excellent suspension).
It would doubtless be reinforced by official crash tests and, even more, by videos of astonishing crash performance vs. steel vehicles. We are aware of no
evidence that weight per se is a customer requirement for light vehicles, and customers who use carbon-fiber sporting goods already know better (pp. 57–60).
577. Toyota’s 2004 lineup has the best average car, truck, and overall fuel economy of any full-line manufacturer in the U.S. market, and includes eight of the
20 most efficient models. Unlike some U.S. competitors, Toyota has met CAFE standards throughout its history. Of 2004 Toyota/Lexus vehicles, 100% are
California-certified as Low Emission (LEV), 75% as Ultra Low Emission (ULEV). Toyota’s 23 models on a leading environmental group’s “Best of 2004” list
(www.aceee.org) top the industry. Thus Toyota’s claim “Cleaner, Leaner and Greener” (Toyota 2004a) is not mere rhetoric.
578. Parker & Shirouzu 2004.
132
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
The endangered automotive sector—why it’s important to act now: Four competitive threats
Implementation
and Toyota Camry. Japanese automakers seek to gain first-mover advantage for the next generation of efficient mobility through experience,
scale, and brand positioning gained with their hybrid vehicles. Korean
firms appear to have similar goals. European automakers, including
DaimlerChrysler, are also backing cleaner diesels, based on the EU regulatory environment and European customer preferences, though their
acceptability under tightening U.S. fine-particulate regulations is not yet
established. While GM is making a major long-term play in fuel cells and
is also exploring and partly adopting other technologies, it may lack the
financial strength to make significant investments into multiple advanced
powertrains, so with only a few mild hybrids during 2004–07, it risks
losing market share to Japanese hybrid-makers meanwhile. Only Ford is
introducing a “strong hybrid” SUV in 2004, later than hoped, under
Toyota technology licenses, and probably using technologies somewhat
less advanced than Toyota’s latest market offerings. In short, U.S. automakers are trying, but are still playing catch-up.
14: Opening moves: Boeing’s bet on fuel efficiency as
the future for commercial aircraft
In chess, the opening moves define the strategy each player will use to prevail. In the transportation
business, the product lines, manufacturing, and marketing approaches define the competitive strategy. The battle between Boeing and Airbus for dominance in civilian aircraft presents a clear example of the challenge and opportunity in using efficient transportation platforms for competitive
advantage, as the combatants have developed opposite strategies.
Airbus entered the commercial aircraft business in the 1990s, and by 2000 had eliminated Boeing’s
near-monopoly, capturing 40% share. Airbus then announced the behemoth A380 , a 555-passenger
plane designed to move passengers between the major international hubs. Airbus has already
received orders for 129 planes. Many major airports can’t yet even accommodate a plane that big.
Boeing instead chose a different approach—its 7E7 Dreamliner—to create business value for the
airlines by boosting fuel efficiency 15–20%. As luxurious as A380 , 7E7 has improved engines, lighter
materials, and better aerodynamics (though of course fewer seats). Boeing expects the airlines to
fly international travelers more frequently along city pairs, in a similar model to the domestic carrier
Southwest. Boeing received its first order for 59 planes (50 from ANA), with more coming in.
As noted in Fortune, “The ideal carrier for Boeing’s new 7E7 hasn’t been invented yet.” But the
reality of the airline industry is that it suffers from overcapacity and poor financial health. Airlines’
difficulty in passing through high fuel costs to their customers and in managing hub congestion
favors Boeing. If Boeing is successful in using 7E7 to reverse its competitive fortunes vs. Airbus,
it will serve as a parable for the whole transportation sector. The countermoves have already begun:
Airbus perceives 7E7 as enough of a threat to its A330 to be contemplating an A350 with a more
efficient engine (p. 158).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
133
Implementation
The endangered automotive sector—why it’s important to act now: Four competitive threats
Japanese automakers
are entering the
carbon-fiber airplane
business with the
clear intention of
cross-pollinating,
so ultralight materials
and manufacturing
methods can start in
the airplane business,
where they’re more
valuable, then
migrate back into
the automotive sector
at higher volumes.
Over time, the heavy
steel cars that now
dominate American
manufacturing and
policy are fated to
occupy a shrinking
global niche.
A deeper competitive threat comes not from innovative propulsion but
from autobody materials. The European Union’s policymakers aren’t yet
committed to a wholesale transition to ultralight materials, but they’re
quietly backing R&D and a supportive policy framework for light-but-safe
cars. European expertise in Formula One racing and some European
automakers’ moves toward carbon composites, especially in Germany,
seem bound to shift EU attention in this direction. Despite similarly
limited policy vision in Japan so far, Japanese automakers are entering the
carbon-fiber airplane business with the clear intention of cross-pollinating,
so ultralight materials and manufacturing methods can start in the airplane business, where they’re more valuable, then migrate back into the
automotive sector at higher volumes. Over time, the heavy steel cars that
now dominate American manufacturing and policy are fated to occupy a
shrinking global niche. In materials, propulsion, and overall design, they’ll
increasingly fit only one country’s market (albeit a huge one), threatened
from all sides. Increasing monetization or regulation of carbon emissions
in key global markets, increasing risk of costly or disrupted oil supplies,
and other countries’ rising attention to both these issues (p. 167) further
heightens U.S. automakers’ prospective competitive disadvantage.579
579. More than 60% of
2002 light-vehicle sales
worldwide were in countries (including China) that
have ratified the Kyoto
Protocol, and some of the
rest, including the U.S.,
are moving in a similar
direction via private market-makers or other forms
of national or subnational
regulation. The Big Three
face greater CO2 exposure
than six of the world’s
seven other major
automakers (all but BMW).
Less because of their product mix than because
they’re behind Toyota in
strong hybrids, GM and
Ford could need bigger
product shifts to deliver a
superior efficiency value
proposition, and those
shifts could become even
more awkward if oil
disruptions required rapid
adjustments (Hakim 2004e;
Austin et al. 2004).
134
An even more subtle competitive threat is in neither propulsion systems
nor body materials, but in vehicles’ systems architecture. At the 1998 Paris
Auto Show, Professor Daniel Roos, who directs MIT’s International Motor
Vehicle Program and coauthored The Machine That Changed the World,
warned automotive CEOs that in two decades most of them could well
be out of business—put there by firms they don’t now consider their
competitors:
In the next 20 years, the world automotive industry will be facing radical change
that will completely alter the nature of its companies and products….In two
decades today’s major automakers may not be the drivers of the vehicle industry; there could be a radical shift in power to parts and system suppliers.
Completely new players, such as electronics and software firms, may be the real
competitors to automakers.
That future is already emerging. And it could encourage formidable new
market entrants. For the software-rich cars of the future, competition will
favor not the most efficient steel-stampers but the fastest-learning system
integrators and simplifiers—manufacturers like Dell and system companies like Hewlett-Packard or Sony. That sort of competition, too, favors
leapfroggers. Most of China’s 300 million cellphone owners never had a
landline phone; they jumped directly into wireless. Most young Chinese
routinely use digital media but have never seen an analog videocassette.
To them, a software-rich, all-digital, all-by-wire vehicle will seem natural,
while a customary U.S. vehicle, mechanically based with digital displays
grafted on, may feel clunky and antique.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
The endangered automotive sector—why it’s important to act now: Four competitive threats
Perhaps the greatest peril, though it’s only starting to emerge on the fringes
of automobility, is transformation of the business model. Traditionally,
automakers sell vehicles and oil companies sell gallons. Both want to sell
more, while customers prefer on the whole to buy fewer (but more physically and stylistically durable) vehicles that use fewer gallons. These opposite interests don’t create a happy relationship. But suppose an automaker
or an oil company leased a mobility service (Box 20, p. 196) that provides
vehicles or other means of physical or virtual mobility, tailored to customers’ ever-shifting needs. Customers would pay for getting where they
want to be, not for the means of doing so. Then vehicles and gallons,
instead of being the providers’ source of profit, turn into a cost: the fewer
vehicles and gallons it takes to provide the mobility service that the customer is paying for, the more profit the provider makes and the more
money the customer saves. Most major car and oil companies are thinking
quietly but seriously about this business model. Most industry strategists
fear that the first firm to adopt it on a large scale could outcompete all the
rest, both because of a better value proposition and because aligning producers’ with customers’ interests tends to yield better outcomes for both.
There is no guarantee that such transformative business models will start
in America. So far, they’ve tended to emerge in small countries, such as
Switzerland and Holland, and in developing countries—that is, in places
where congestion and high fuel and land prices force innovation at the
most fundamental level, transcending mere technology.
China and India
Preoccupied with obvious competitors in Japan, Korea, and Europe,
American automakers risk being blindsided by China’s automakers.
The Chinese are positioning themselves to enter the world market with a
leapfrog play of highly efficient vehicles, backed by the strength of the
scale and experience gained in their burgeoning 4.3-million-vehicle/y
(2003) domestic market. The Chinese government’s National Development
and Reform Commission’s 2 June 2004 white paper steers the $25.5 billion
automaking investment that’s expected during 2004–07 to raise output to
nearly 15 million vehicles a year, nearly as big as the whole U.S. market.
This policy is also raising barriers to market entry (each project must
invest at least $241 million including $60 million of R&D), consolidating
the fragmented 120-plant automaking sector, and tightening domestic
efficiency standards that most heavy U.S. vehicles can’t meet (p. 45).
Hybrids and lightweighting are specifically to be encouraged.580
Moreover, this automotive strategy is intimately linked with a transformative energy strategy. Four weeks after its release, Premier Wen
Jiabao’s Cabinet approved in principle a draft energy plan to 2020.
It makes energy efficiency the top priority, pushes innovation and
advanced technology, and emphasizes environmental protection and
energy security to “foster an energy conservation-minded economy
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
It’s dangerous to
underestimate the
dynamism of Chinese
manufacturing.
Chinese industry is
supported by a strong
central policy
apparatus rooted in
five millennia of
history, and is
enabling the world’s
most massive
construction boom
in at least
two thousand years.
Integrated advances
in China’s policies
for both cars and
energy will probably
make its auto
industry a major
global competitor in
the next decade, with
India following fast.
In a few decades,
a mighty Chinese
economy’s automakers might have
taken over or driven
out the Big Three
and even bought
major Japanese
automakers.
580. APECC 2004.
135
Implementation
The endangered automotive sector—why it’s important to act now: China and India
There’s also India’s
younger, smaller,
but rapidly maturing
auto industry.
It’s only $5b/y, but is
growing by 15% a
year. Quality has
improved so rapidly,
on a Korea-like
trajectory, that in
2004, Tata Motors is
exporting 20,000 cars
to the UK.
and society” in which “the mode of economic growth should be transformed and a new-type industrialization road taken.”581 As other industries have learned the hard way, WTO-governed global competition
works both ways, and it’s dangerous to underestimate the dynamism of
Chinese manufacturing. Chinese industry is supported by a strong central
policy apparatus rooted in five millennia of history, and is enabling the
world’s most massive construction boom in at least two thousand years.
Already, too, homegrown Chinese fuel-cell cars are rapidly advancing in
several centers, raising the likelihood that Chinese leaders’ aversion to the
oil trap will be expressed as leapfrog technologies not just in efficient
vehicles but also in oil-free hydrogen fueling. In a few decades, a mighty
Chinese economy’s automakers might have taken over or driven out the
Big Three and even bought major Japanese automakers.
Unusual U.S.
conditions—
cheap gasoline,
high incomes,
weak efficiency
regulation—
don’t prevail in most
global markets, and
blunt the Big Three’s
competitive edge.
581. APECC 2004, p. 5
582. Farrell & Zainulbhai 2004.
583. Prahalad 2004.
136
In case 1.4 billion Chinese moving rapidly to make something that beats
your uncle’s Buick isn’t enough of a threat, there’s also India’s younger,
smaller, but rapidly maturing auto industry. It’s only $5b/y, but is growing by 15% a year. Quality has improved so rapidly, on a Korea-like trajectory, that in 2004, Tata Motors is exporting 20,000 cars to the UK under
the MG Rover brand. India “may be better placed than China is to
become a global low-cost auto-manufacturing base.”582 A billion Indians,
with an educated elite about as populous as France, have already transformed industries from software to prosthetics, using breakthrough
design to undercut U.S. manufacturing costs by as much as several hundredfold.583 India’s domestic car market, like China’s, is evolving under
conditions that favor an emphasis on fuel efficiency.
Suppressing the signals
With such competition looming, one might suppose that U.S. automakers
would embrace tighter domestic efficiency requirements to help them
gird for the challenge. Instead, their visceral and partly understandable
revulsion to regulation led them to invest heavily in full-court-press lobbying to freeze or weaken current requirements. That lobbying’s success
could set the stage for a rerun of Japanese competitors’ 1970s market success, but this time against far more competitors than just Japan.
Historically low U.S. fuel prices, high personal incomes, and low and
stagnant U.S. efficiency standards all encourage U.S automakers to undercompete globally on fuel economy. None of these three peculiarly North
American conditions prevails in most of the rest of the world. The disparity could get worse. If the National Highway Transportation Safety
Administration adopts its spring 2004 proposal to base CAFE standards
on weight class—deliberately rewarding heavier (except the very heaviest) and penalizing lighter vehicles—it will enshrine the already worrisome divergence between U.S. and other, especially European, safety
philosophies, making it ever harder for U.S. automakers to market their
heavy products abroad (p. 58).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
The endangered automotive sector—why it’s important to act now: Suppressing the signals
Implementation
Clinging to heavy and inefficient product lines will make the U.S. the
If the U.S. fails to act decisively
unrivaled leaders in producing products that the world’s consumers do
and coherently during this period,
not want, in commanding the loyalty of a steadily aging and shrinking
we will not only continue to lose
customer base, and in depending on perpetually cheap fuel that seems
market share: we will ultimately
ever less likely to remain reliable.
lose the manufacturing base, and
those high-wage manufacturing
jobs will be lost forever. Clinging to heavy and inefficient product lines
will make us the unrivaled leaders in producing products that the world’s
consumers do not want, in commanding the loyalty of a steadily aging
and shrinking customer base, and in depending on perpetually cheap
fuel that seems ever less likely to remain reliable. Worse, we will fall
behind in the advanced materials technology race, ceding the next generation of high-technology manufacturing jobs to agile and uninhibited
competitors overseas. Add the subtler forms of competition in softwaredominated vehicles and solutions-economy business models, and the
prospects dim further for automakers that perpetuate existing product
lines, changing only incrementally, cocooned within a comfortable regulatory and price environment.
Crafting an effective energy strategy:
transformative business innovation
Detroit’s dilemma is unpalatable: if countries like China and India do
leapfrog to ultralight fuel-cell cars, they’ll pose a grave competitive threat
regardless of oil price, but if they don’t, their growth in oil demand,
and the cashflow needs of demographically and socially challenged oil
exporters, will conspire to keep upward pressure on fuel prices more
consistently than in the 1990s, making customers even more dissatisfied
with Detroit’s slow progress toward oil-frugal cars. For the U.S. economy,
either path—importing cheaper superefficient cars or importing costlier
oil—creates a drag on growth, drives inflation and interest rates, and
destroys good jobs.584 When thoughtful auto executives say they’re
uncomfortable having their industry’s future depend on unpredictable oil
prices, they’re right—but in a far deeper sense than they may intend.
The imperatives of the end of the Oil Age converge with those of radically
shifting automotive technology to compel the linked transformation of
these two largest global industries. If their twin transformations dance
smoothly together, it’ll be sheer delight. But if either industry lags,
it’ll get stepped on, and more agile partners will cut in.
U.S. automakers
must shift speedily
to high efficiency,
then beyond oil—
or risk suffering the fate
of the once-mighty
whaling industry.
584. IEA 2004b. From 1970 to 2003, inflation has historically closely tracked the price of oil. U.S. GDP growth has moved
inversely with inflation. While the shift in manufacturing and services mix has lowered the direct impact of oil on the U.S.
economy, and so has reduced oil intensity, the general vulnerability of the economy to inflationary pressures remains.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
137
Implementation
“Disruptive
technologies”
can make rapid
vehicular innovation
and industrial transformation the path to
success, re-equipping U.S. automakers
to compete and prevail in the global marketplace—or can
spell their demise if
others do it first.
The Big Three are
slowly being forced
up the value chain
by the foreign
competition, and may
soon find themselves
stranded without
much of a market.
585. Schumpeter
1943/1997.The term probably dates to 1942.
586. Schumpeter 1943/1997.
587. Schumpeter 1943, p. 31.
588. For a discussion of
creative destruction
applied to the telecommunications sector and the
impact of the Internet on
industries, see McKnight,
Vaaler, & Katz 2001.
589. Christensen 1997.
590. Henley 2004;
Power & Wrighton 2004;
Naillon 2004.
591. Austin et al.
(2004, p. 11) summarizes
the strategic trap of U.S.
automakers’ excessive
light-truck dependence.
138
Crafting an effective energy strategy: transformative business innovation
The creative destruction dilemma
More than 50 years ago, economist Joseph Schumpeter described the
process of “creative destruction,” where innovations destroy obsolete
technologies, only to be overthrown in turn by ever newer, more efficient
rivals.585 Creative responses to economic shifts occur when businesses
respond to innovation in ways that are “outside the range of existing
practice….Creative responses cannot be predicted by applying the ordinary rules of inference from existing facts, they shape the whole course of
subsequent events and their long run outcome.”586 Depending on how
radical the change is and how flexible the sector is, the impetus for
change can often come from within the industry. In Capitalism, Socialism,
and Democracy, Schumpeter argued in 1943 that businesses are “incessantly being revolutionized from within by new enterprise, i.e. by the intrusion of new commodities or new methods of production or new commercial opportunities into the industrial structure as it exists at any
moment.”587 This process has played out across many industries, most
recently those transformed by the Internet. Now, focused partly by the
concerns about oil and climate, it has come squarely to the transportation
sector.588 How can this sector’s existing players address the creative
destruction challenge?
In The Innovator’s Dilemma, Harvard Business School professor Clayton
Christensen explained how industry leaders get blindsided by “disruptive innovations” because they focus too closely on their most profitable
customers and businesses.589 The U.S. car and truck companies’ asset base,
supplier networks, and labor contracts are tied to manufacturing highly
profitable but inefficient light trucks and SUVs. Their business focus is
on how to provide larger, more powerful, and higher-margin vehicles to
their most profitable customers in this segment, virtually ignoring the
underserved, low-margin small-car segment that has been scooped up
by the new Korean entrants Hyundai and Kia. (The Big Three have even
come dangerously close to ignoring the entire car sector.) The increasing
difficulty of competing in small and medium cars heightens dependence
on SUV profits. However, those same SUVs are too large and too fuelconsumptive to gain a market foothold in countries outside the U.S.; the
French government has proposed stiff taxes on them, with growing support from other European countries, and the government of Paris is even
considering outlawing them as a public nuisance.590 So while the upsurge
in U.S. demand for SUVs was an unusually persistent bonanza for Detroit
because competitors at first thought SUVs were a passing fad, the global
market share of the Big Three continues to shrink.591 They’re slowly being
forced up the value chain by the foreign competition, and may soon find
themselves stranded without much of a market. Their dilemma is akin to
the Swiss watchmaking industry’s plight before Nicholas Hayek’s radical
simplifiers at Swatch saved their market from Asian competitors—starting with the cheapest commodity watches.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: transformative business innovation: The creative destruction dilemma
In The Innovator’s Dilemma, Christensen observes that the successful incumbents are able to adapt to evolutionary change with what he describes as
“sustaining innovations”—those that make a product or service better in
ways that customers in the mainstream market already value, like the low
direct operating cost of Boeing’s new 7E7 (Box 14).592 The SUV itself was a
sustaining innovation for the U.S. car companies, because it was a breakthrough product that could be effectively marketed as offering the automotive industry’s best customers more perceived value than before.
Within this framework, the superefficient automobiles discussed on pp.
44–72, together with the new business models they offer (in product line,
distribution, sales, aftermarket),593 are disruptive innovations like the ones
Christensen describes. Disruptive innovations create an entirely new market by introducing a new kind of product or service or by enabling new
markets to come into being.594 (The vehicles in our State of the Art portfolio
are especially disruptive innovations because unlike their Conventional
Wisdom brethren, they enable and accelerate the shift to hydrogen fuel-cell
vehicles and hydrogen as an entirely new energy carrier, as we’ll see on
pp. 233–234.) The problem is that incumbent companies, no matter how
successful, have trouble managing or initiating disruptive changes to
their own markets. To see why, we must understand the business challenges faced by the automotive industry.
Business challenges:
market, business, and customer realities
Most U.S. light-vehicle buyers scarcely value fuel economy
Businesses and their customers appear to care deeply about energy only
when it is not available. Scarcity or disruption of supply creates far more
market reaction than an increase in market prices.595 The response to the
1970s oil shock was a dramatic shift by customers to more efficient vehicles—often this meant smaller vehicles because choices at the time were
limited, although, as noted on p. 7, this shift to smaller (mainly imported)
cars is actually an urban legend. In fact, only 4% of the 7.6-mpg newdomestic-car efficiency gain during 1977–85 came from making cars
smaller.596 The shift to efficient vehicles, coupled with the political will to
Implementation
As then United Auto
Workers President
Doug Fraser noted in
1980, “If the 1975
[CAFE] legislation
had not been enacted,
there quite possibly
could have been
even greater damage
inflicted on the
industry and its
workers by
foreign imports.”
Market conditions
will continue to
retard or block
adoption of superefficient light vehicles,
unless new business
strategies and public
policies shift buyers’
behavior.
U.S. autobuyers
count only the
first three years’
fuel savings—
a ~58% underestimate.
592. Christensen & Overdorf 2001, p. 114.
593. For example, the wireless services mentioned on p. 63. Smaller-scale manufacturing would permit greater localization and a sales model like that of a
Dell customized mail-order computers with onsite General Electric service contracts. (This is a sales model Ford in the UK introduced for servicing conventional cars as a convenience dimension; Ford then found it was also cheaper to deliver than traditional drive-in service shops.) Models will vary widely
between cultures—most cars in Japan, for example, are sold door-to-door, often through longstanding relationships, and usually bundled with all the
required financial and legal services—but we believe the novel technical characteristics of software-rich ultralight vehicles could transform their market
and aftermarket structure as much as their manufacturing.
594. Christensen & Overdorf 2000.
595. Steiner 2003. For example, during the height of the 2000–01 U.S. power crisis, when prices soared, over 60% of business users rated reliability as their
most important concern, as compared to prices, which were a distant second (Datta & Gabaldon 2001).
596. See the detailed finding of only a 0.5-mpg gain from 1975–93 shifts in size class (Greene & Fan 1995). Interestingly, although CAFE standards didn’t make
vehicles significantly smaller, gasoline price spikes have historically done so (Greene 1997, p. 26 & Fig. 10), especially for light trucks.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
139
Implementation
Crafting an effective energy strategy: Business challenges: Most U.S. light-vehicle buyers scarcely value fuel economy
Buyers underestimate
by ~58% the true fuel
economy benefits of
a typical light vehicle
over its 14-year
average life, because
they count only ~3 of
its 14 years of fuel
savings—not enough
to repay the extra
cost of a more
advanced and
efficient model.
enact CAFE standards and Detroit’s impressive response to them, led to
a whopping 60% increase in new light vehicles’ fuel economy in the
decade 1975–84.597 At the time, most in Detroit were not enthusiastic about
this legislative mandate, and many vehemently opposed it, but as then
United Auto Workers President Doug Fraser noted in 1980, “If the 1975
[CAFE] legislation had not been enacted, there quite possibly could have
been even greater damage inflicted on the industry and its workers by
foreign imports.”598 Yet today, most industry observers believe that U.S.
customers will continue to buy SUVs until gasoline prices reach $3–4/gal,
consistent with experience in Europe where $3–5/gal fuel holds SUVs’
market share to one-fifth the U.S. level.599 What has changed since the oil
price shocks of the 1970s?
597. EIA 1994.
598. DeCicco, Griffin, &
Ertel 2003, p.6.
599. Hakim 2004d.
600. Hakim 2004d. By 1998,
the share of household
income paid for energy use
was comparable to the
share paid in 1973, before
the oil prices for most IEA
countries soared (IEA 2004).
601. Stern et al. 1986. An
energy efficiency project
with a two-year simple payback and a five-year project life yields a 41%/y IRR.
602. Ball 2004.
603. NAS/NRC 2002a.
Honda of America similarly
reports that consumers
only value fuel savings for
the first 50,000 miles of the
a car (German 2002) More
broadly, the U.S.
Department of Energy survey found that average
consumers need to be paid
back within 2.9 years for an
investment in fuel economy
(Steiner 2003)
A big difference is that energy accounts for only 4.5% of the average
American’s disposable income today, compared with 8% during the 1970s
oil shocks. Both disposable income and credit have increased, so higher
energy prices don’t pinch most households’ budgets—until a price spike
occurs.600 Most people can’t be bothered to buy energy efficiency as insurance against price spikes. U.S. households typically require paybacks within two years to invest in energy-efficient appliances, an implicit discount
rate over 40%/y—an order of magnitude higher than a social discount
rate, and several times even a credit-card cost of capital.601 For the lowerincome families on which energy price spikes inflict real hardship,602 lack
of credit and disposable income increases this discount rate substantially,
and energy takes a higher fraction of household income but there’s less
money to invest in efficiency. Thus, the tendency to underinvest in energy
efficiency prevails across income levels. Moreover, it applies to vehicles as
well as in the home: most vehicle buyers count only the first three years’
energy savings.603 The implication is that buyers underestimate by ~58%
the true fuel economy benefits of a typical light vehicle over its 14-year
average life.604
There is hope. One-fifth of U.S. light-vehicle buyers say they are early
adopters, willing to spend $2,500 more for better fuel economy.605 That’s
about what a State of the Art SUV would cost extra, or as much as a hybrid
does today. This early-adopter segment may be large enough to kick-start
the market (albeit with a gentle kick), as evidenced by hybrid cars’ recent
rapid sales growth. In 2004, the problem isn’t inventing attractive doubledefficiency vehicles like today’s hybrids (which save nearly twice as much
fuel as Conventional Wisdom vehicles would), but being unable to make
them quickly enough, so Toyota is already planning to build an additional
hybrid-car factory.
Based on historic purchasing behavior, though, only a minority of customers will spontaneously adopt a hybrid that saves gasoline at a cost
comparable to today’s gasoline price, or even a State of the Art vehicle that
604. Greene et al. 2004.
605. Steiner 2003.
140
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business challenges: Most U.S. light-vehicle buyers scarcely value fuel economy
saves gasoline at a cost of 56¢ a gallon. Many more will buy an incremental Conventional Wisdom vehicle that saves 61% less fuel than State of the
Art but at a 71% lower cost per gallon (15¢/gal). Such a vehicle, with an
extra cost of only $727, is more attractive to most car buyers because they
count only ~3 of its 14 years of fuel savings—not enough to repay the
extra cost of a more advanced and efficient model. But once they’ve
bought an incrementally improved vehicle, fewer State of the Art vehicles
will be bought for the next ~14 years, slowing the fleet’s shift to really
high efficiency. The market reality is that most customers are slow to
adopt very efficient vehicles or appliances, and it takes even longer for
those new purchases to become a major fraction of the whole fleet on the
road. But the sooner that adoption starts, the sooner it affects the entire
fleet. We’ll therefore suggest, starting on p. 178, creative ways to jumpstart early adoption by customers and correspondingly early retooling by
automakers.
Most firms underinvest in energy efficiency too
Even sharp-penciled businesses underinvest in energy efficiency, despite
the exceptionally high rates of return on these investments.606 The discount
rate for U.S. business investments in new capital projects is generally
estimated at ~15%/y, but investments in energy efficiency projects that
are often taken from operational, rather than capital, budgets commonly
require a 2-year payback, implying a 41%/y real discount rate if the
project has a 5-year life, or 50% over 20 years—around ten times the cost
of capital. Many firms nowadays ration capital, despite its record-low
cost, so stringently that they require energy efficiency to repay its cost in
a single year! There appear to be at least five main reasons for this egregious misallocation of capital: 607
Implementation
Five well-known
organizational failures
lead most companies
to misallocate capital
away from very
lucrative, low-risk
investments to cut
energy costs, including
those of vehicles.
• The capital budgeting process in most corporations not only allocates
capital to the projects with greatest return, but tends to skew capital
towards projects with greater growth prospects—those that gain markets rather than cut costs. Hence, new plants will often be built before
the old ones are made more efficient.
• Many firms fail to risk-adjust competing investments’ returns.
They therefore fail to notice that competent investments in energy efficiency are among the lowest-risk opportunities in the whole economy;
indeed, they’re a bit like insurance, in that their energy savings are
worth the most when they’re most needed because high energy prices
have cut revenues and profits. Investment to gain or keep market share
is far riskier than, say, buying an efficient vehicle that is certain to save
a rather well-defined amount of fuel and to reduce proportionately
the firm’s exposure to fuel-price volatility. A firm that takes financial
economics seriously will therefore buy efficiency whose returns go
right down to, and even a bit below, its marginal cost of capital.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
606. See review and
citations in Lovins & Lovins
1997; DeCanio 1993;
DeCanio 1998; Howarth &
Andersson 1993;
Koomey, Sanstad, & Shown
1996; Lovins 1992.
607. Golove & Eto 1996;
Koomey 1990.
141
Implementation
Five main reasons
for misallocation
of capital
(continued):
Crafting an effective energy strategy: Business challenges: Most firms underinvest in energy efficiency too
• Managers tend to think of energy costs as too small a portion of the
overall cost structure, and too diffuse, to merit attention—even though
saved energy, like any saved overhead, drops straight to the bottom
line, where it can add greatly to pinched net earnings.
• Incentives are often misaligned, so that the party who would save the
money from reduced energy use does not own the energy-using equipment, such as a leased commercial office building or a trucking fleet.
• Lack of activity-based costing and cost accountability often obscure
energy costs from line managers, who may never even see the bills
because they’re sent directly to a remote head office.
Nearly all businesses
fall far short of buying
the economically
optimal amount of
energy efficiency
in their equipment
and vehicles.
So it should come as
no surprise that
manufacturers are
reluctant to invest in
retooling their product lines to produce
more efficient cars,
trucks, and airplanes.
Once again, the market reality is that nearly all businesses fall far short
of buying the economically optimal amount of energy efficiency in their
equipment and vehicles. The organizational reasons are complex, but the
reality is widely known and is accepted by all knowledgeable analysts.
If we want to change that behavior, we must change the underlying conditions that cause it. Otherwise it should come as no surprise that given
such customer indifference, manufacturers are reluctant to invest in
retooling their product lines to produce more efficient cars, trucks, and
airplanes. It simply looks too risky to make big investments to make new
products that customers may not buy.
The investment required to launch a new automotive product line can be
around $1–2 billion.608 Boeing and Airbus are spending $10 and $12 billion,
respectively, to develop their newest civilian airplane platform.609 Truck
product line development costs ~$0.7–1 billion.610 For the transportation
sector, therefore, developing a major new product line is a daunting risk,
even a bet-your-company decision. That’s why Ford, for example, chose
to add hybrid powertrains to existing product lines (reportedly making the
hybrid Escape potentially profitable 611) rather than design a new hybrid
platform from the ground up.
Having already placed some bets on hybrids (mainly Ford so far) and
fuel cells (chiefly GM and DaimlerChrysler, each to the tune of $1 billion),
U.S. automakers no longer have the financial strength to shift to ultralight
autobodies at the same time. And they may worry that if they did place
that bet, they’d miss the opportunity to edge out their rivals by incremental improvements in conventional markets that they thoroughly understand. Given today’s market conditions and torpid policy environment,
this reaction is superficially plausible. And it’s reinforced by the volatile
way capital markets respond to energy prices.
608. For example, the Saturn cost $2 billion to launch.
DaimlerChrysler spent over $2 billion developing its lineup
of mini-vans (Greenberg 2001). Ford even spent $6 billion
on the Mondeo, the first “World Car” (Law 2004).
142
609. Bloomberg.com 2004a; Nelson.com, undated.
610. Thomas 2002.
611. Parker 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business challenges: Most firms underinvest in energy efficiency too
Capital markets are notoriously fickle, with some reason, and most of
all in the feast-and-famine world of energy overshoots. Many leading
investors today still bear thick scar tissue from the mid-1980s, when public policy pushed hard to expand energy supplies at the same time that
earlier policies and price signals from 1979’s second oil shock invisibly
but dramatically increased end-use efficiency. These two trends met headon in 1985–86 as efficiency captured the market sooner, shrank the revenues that were supposed to repay the supply investments, glutted energy markets, crashed prices, and bankrupted many suppliers. (Those with
short memories periodically try to persuade investors to rerun this bad
movie, generally reaching the same sad ending.) Herd-instinct investors,
who lack a long view and steady nerves, tend to make energy technology
stocks soar after disruptive events. These stocks then tend to crash as
energy markets stabilize.612 Less seasoned investors who haven’t experienced the painful reality of market equilibration amplify these extremes
by behaving as if only increased supply, not increased efficiency, is real,
reliable, and bankable. Ignoring efficiency can be terminal.
The risk of being risk-averse
No wonder many analysts openly doubt if consumers will adopt hybrid
cars in enough volume to justify even the modest investments made by
Ford, Toyota, and Honda. Although rapid market adoption so far seems
to defy cynicism,613 a policymaker for a U.S. automaker even remarked in
mid-2004 that Prius “has not been accepted by the American public.”
Ironically, Toyota was simultaneously announcing a near-doubling of
Prius production and considering building a U.S. plant to make more,
Prius was flying off dealers’ lots faster than any other car in America for
the tenth straight month (nearly twice as fast as the runner-up), and some
Prius dealers were extracting price premia greater than the rebates his
company was paying everywhere to sell its flagship products. Doubting
the staying power of a successful rival is easy but imprudent, especially
when it’s as formidable a firm as Toyota.
Large incumbent firms like to stay within their management comfort zone
and to choose either incremental or sustaining innovations. Yet business
history teaches us that no matter how big or old the company, this strategy will ultimately succumb to the onslaught of innovative competitors.
Disruptive innovations, and the companies that undergo transformative
change to adopt them, ultimately defeat creatures of habit; in the long
run, standing pat is a losing strategy. However accustomed its customers
and compliant its regulators may be, any big, muscular firm will turn stiff
and sluggish without frequent stretching. The market says so. Competing
against aggressive innovators, Ford and GM rank number 3 and 4 in
revenues among the Fortune 500 companies, but 80th and 87th in market
capitalization.614 As we noted on p. 30, their combined market capitalization plus DaimlerChrysler’s is less than Toyota’s. Moreover, 95% of GM’s
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
It’s easy but dangerous to be complacent
about competitors’
progress. Winning
depends on strategy
as well as tactics.
Business history
teaches us that no
matter how big or old
the company, the
strategy of incremental
innovations will
ultimately succumb to
the onslaught of
innovative competitors.
Disruptive innovations,
and the companies
that undergo transformative change to
adopt them, ultimately
defeat creatures of
habit; standing pat is
a losing strategy.
612. Datta & Gabaldon 2003.
613. Golfen 2004.
614. Hakim 2004b.
143
Implementation
Crafting an effective energy strategy: Business challenges: The risk of being risk-averse
market capitalization is based on value from existing assets, rather than
new growth opportunities: apparently the market doesn’t believe GM has
the capability to break out of incremental innovation.615 U.S. automakers
rank near the bottom in return on equity, and often make more money
from financing than from manufacturing vehicles.616 All who know firsthand these firms’ extraordinary technical capabilities and their pivotal
role in the U.S. economy hope for a magic turnaround, but Disney’s First
Law notwithstanding, wishing won’t make it so.
Market capitalism,
the most powerful
engine of wealth
creation the world
has ever known,
bears the seeds of
its own renewal.
Financial weakness is rooted in, and reinforces, underlying organizational
flaws. Christensen observes that incumbent organizations have tremendous inertia due to their capabilities, processes, and values (defined not
as ethics, but as “the standards by which an employee sets priorities to
judge what [business opportunities] are attractive or unattractive”).617
McKinsey’s Foster and Kaplan note a similar phenomenon they call “cultural lock-in,” wherein companies maintain their old mental models and
are unable to move beyond incremental innovation.618 We recognize, and
sympathize with, the formidable leadership and management challenge
that the Big Three and their associated business network face in trying to
inspire and reward breakthroughs in a culture of caution—one where it’s
definitely easier to get permission than forgiveness, and (as the Japanese
proverb puts it) the nail that sticks up gets hammered down. But we
believe that a combination of leadership from within, integration of new
technologies, and public policy reinforcement from without holds a timely answer to this challenge.
Milton Friedman, viewing the enormous capacity of the business system
to block change, once remarked that the problem with capitalism is capitalists. (He added that the problem with socialism is socialism.) Yet
through the harsh and necessary discipline of creative destruction, market
capitalism, the most powerful engine of wealth creation the world has
ever known, bears the seeds of its own renewal.
Ford and GM rank number 3 and 4 in revenues among the Fortune 500 companies,
but 80th and 87th in market capitalization—
their combined market capitalization
plus DaimlerChrysler’s
615. Christensen & Raynor 2003.
is less than Toyota’s.
616. The joke around Detroit is that the Big Three make
cars so they can loan people money to buy them. Maybe
it’s not a joke. In 2003, Ford Motor Company’s Financial
Services organization posted income (before taxes) of
$3.3 billion while the Automotive business had $0.1 billion
in income before tax and excluding special loss items
(Ford Motor Company 10-K, 12 March 2004). GMAC, the
financing organization of General Motors, posted $2.8 billion of net income vs. GM’s automotive net income of $0.6
billion (General Motors Corporation 10-K, 11 March 2004).
617. Christensen & Overdorf 2000.
144
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: transformative business innovation
Business opportunities:
competitive strategy for profitable
transformation in the transportation sector
In The Innovator’s Solution, Clayton Christensen and Michael Raynor discuss how incumbent companies can create disruptive innovations rather
than be destroyed by them. They argue for a managerial approach that
demands a focus on disruptive innovation, and for the fundamental
changes in the internal business processes and values that are needed to
reshape an organization. McKinsey’s Foster and Kaplan argue for the
same focus on disruptive technologies, but choose a different managerial
solution that focuses on adaptive processes and the decisiveness of private equity managers. Both concur on the fundamental principle that the
innovation focus should be on disruptive, not sustaining or incremental,
innovations—and that management must start this effort when the current business is in its prime, which is far earlier than most managers
expect. By the time a business is mired in low-margin, commoditized
incrementalism, it no longer has the strength and agility to break out of
the swamp, and falls easy prey to determined predators.
The business cases detailed in Technical Annex, Ch. 20, and summarized
next, show that the transition to State of the Art cars, trucks, and airplanes
will be profitable for manufacturers, particularly as the normal and
expected technological progress continues to pare their manufacturing
investment risks. Since the superefficient technologies offer new and better value propositions for end-users (pp. 61–73), producers must develop
better production techniques and business models to manufacture and
deliver them, as we discuss next, and should welcome and seek supportive public policies to ease their manufacturing conversion (pp. 178–207).
America’s transition to radically more efficient light vehicles, in particular,
is of historic dimensions. It will deliver a 64–78-plus-mpg SUV (for example) to customers, at an attractive price, without compromising any customer
attribute. This will occur fast enough and soon enough only if business
and investment risks are lowered so that the perceived benefits of shifting
corporate cultures and strategies will exceed the perceived hazards.
The business risk can be lowered if companies stimulate customer demand
without shifting preferences between vehicle types or sizes; we’ll describe
on pp. 178–203 how to do this. The investment risk can be lowered if
companies take a new approach to developing and manufacturing highefficiency vehicles. This requires two breakthroughs: new manufacturing
processes that lower the capital investment, and significantly improved carboncomposite (or lightweight-steel) manufacturing processes to cut variable production costs (materials, fabrication, and finishing). These co-evolving breakthroughs can already be seen in the laboratory and are starting to enter the
market (pp. 56–57).619 Their refinement and scaleup will doubtless require
some trial-and-error, but both these breakthroughs already appear to have
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Superefficient
vehicles offer
superior value
propositions not just
for customers but
also for automakers.
Delivering a
64–78-plus-mpg SUV
(for example) to
customers, at an
attractive price,
without compromising any customer
attribute requires
two breakthroughs—
new manufacturing
processes that
lower the capital
investment
and
significantly improved
carbon-composite
(or lightweight-steel)
manufacturing
processes to cut
variable production
costs . These can
already be seen
in the laboratory and
are starting to enter
the market.
618. Foster & Kaplan 2001.
619. Whitfield (2004) partially
updates the status of two
developments described
above on pp. 56–57.
145
Implementation
Ultralight composite
vehicles can create
breakthrough
manufacturing
advantages.
Assembly plants
can need two-fifths
less capital investment and incur
lower variable costs.
Development cycles
may also become
faster and cheaper.
The business logic
is compelling
for automakers—
and the vehicles
save fuel at far lower
cost than today’s
best-selling hybrids.
Crafting an effective energy strategy: Business opportunities: competitive strategy for profitable transformation
progressed beyond the realm of potential showstoppers, because they
combine in new ways proven techniques that are already successfully
practiced in other contexts, and they draw on a risk-managed portfolio of
diverse methods. So how can these manufacturing breakthroughs cut
the investments, variable costs, and risks of making State of the Art light
vehicles? And how can analogous transformations be accelerated in the
heavy-truck and airplane businesses?
Lowering light vehicles’ manufacturing risk
New automotive platform development is and will remain an extraordinarily costly, demanding, and exacting business. However, its investment
requirement has recently been reduced from an average of $2 billion to
~$1.5 billion. The development cycle has also become much faster, falling
from ~60 to ~18 months, as design and even crash-pretesting shifts from
physical prototypes to the sophisticated on-screen virtual-design and simulation software proven in such aircraft as 777 and B-2.620 Product-development costs and times for State of the Art light vehicles should be no
greater, and may well be smaller, due to unprecedented parts de-proliferation (which lowers the integration complexity), modular but integrated
vehicle operating software, and the greater technical (though not cultural)
simplicity of designing and producing vehicles from advanced composite
materials.621
Manufacturing investment and variable cost
Advanced-composite autobodies can considerably reduce the required
capital investment for building new car manufacturing/assembly plants.
(The steel industry claims competitive production cost for its ULSAB-ATV
concept design, though for different reasons.) As Box 15 explains, with
ultralight composite autobodies, the cost of a 50,000-vehicle/y greenfield
assembly plant could drop to ~$185 million, or $3,700/vehicle-y—about
40% less than the $6,150/vehicle-y at GM’s most modern C-flex plant,622
which in turn is nearly one-fifth cheaper than typical Japanese transplant
facilities.623 The cost reduction is due primarily to the elimination of the
paint shop, and secondarily to far fewer and lower-pressure presses,
cheaper tooling (adjusted to the same lifetime), no welders, and simplified assembly. State of the Art vehicles’ redesign also lowers the minimum
620. Mateja & Popely 2004; Weber 2002; Gould 2004.
621. Proprietary estimates of current product development costs for carbon-composite vehicles obtained from D.R. Cramer (Hypercar, Inc.), personal communication, 22 July 2004, based on diverse 1998–2004 internal analyses.
622. Proprietary estimates of current production costs for a State of the Art vehicle (from Hypercar, Inc., updated in 2004 by Whole Systems Design [Box 9])
are consistent with the $3,700/vehicle-y investment estimate, and the advanced-composite autobody costs are far below those stated by Cramer & Taggart
2002. GM’s 130,000 vehicle/y Lansing Grand River facility (Ward’s Auto World 2002, Automotive Intelligence News 2004), its most modern assembly plant,
cost $800 million but is half the size and cost of its predecessors, and “embodies everything we’ve learned about lean manufacturing,” says Gary Cowger,
president of GM North America (Weber 2002). Production ramp-up reached a total of 59,128 units in 2003 (Garsten 2004).
623. JAMA 2003.
146
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business opportunities: Lowering risk: Manufacturing investment and variable cost
Implementation
15: Radically simpler automaking with advanced composites
A large part of the expense of creating a light
vehicle comes from the dizzying complexity of
its manufacturing process. A typical car can
easily contain some 10,000–15,000 separate
parts or even more.624 Automakers, on average,
have devolved about a third of their total engineering effort to their parts suppliers, and that
fraction is rising, in many cases to two-thirds. In
the manufacturing plant, these thousands of
parts must be coordinated so that there are just
enough parts available, not too many, all defectfree, in exactly the right place at the right time.
Up to about a hundred (sometimes more) different steel body parts are stamped from flat
sheetmetal into final shape in gigantic presses,
using progressive tool-steel die-sets to form
each part gradually through several successive
hits. Those tools cost about a half-billion dollars,
not counting the high-pressure presses they fit
into (p. 73).
Once stamped, the body parts are brought
together and precisely positioned by giant jigs
while hundreds of robots join them together with
~2,000+ spot-welds into the “body-in-white”—
the car’s basic frame, body, and “closures”
before finishing. Then the body is cleaned and
coated by a rust-inhibitor. Painting, the hardest
step, accounts for about half an assembly plant’s
total capital cost (a new paint shop typically
costs ~$100–200 million) and for at least a quarter of the total cost of creating a painted steel
part. After drying and inspection, the painted
body moves slowly down an assembly line
where workers, with some robotic help, add the
other major components such as engine,
drivetrain, suspension, brakes, wheels, wiring,
and interiors.
Advanced-composites autobodies can eliminate
or greatly simplify body manufacturing and final
assembly, slashing both capital and operating
costs. A composite autobody could be made
from 10–20 parts rather than the 60–100 typical
of a modern steel unibody.625 Incoming inventory,
welding, trim, body assembly, and worktime in
final assembly could be reduced. Also, by making autobody parts from advanced composites
molded at modest pressures rather than from
stamped sheet steel, and laying color in the
mold (integral to the polymer matrix or as a dry
film coating), steel stamping operations and
painting could be completely eliminated and
replaced by composite forming equipment,
either with no paint line or with a simpler one
for clearcoating in-mold-pigmented composite
parts. As in-mold color matures further, current
designs, which add shiny cosmetic polymer
exterior panels over the structural carbon-composite passenger cell, might even evolve to a
true “monocoque”: like an egg, the shell is the
structure. But even using current techniques,
the combination of structural and exterior-panel
elements would be simpler to form, assemble,
and finish than today’s steel unibody. And integrated design that emphasizes lightweighting
and parts-reduction not only halves vehicle
weight, making the propulsion system far smaller and cheaper; it can also eliminate or combine
many other current vehicle systems.626
624. A source estimating 20,000 parts in a 1996 car (Life 1996, p. 20) reports that Ransom E. Olds’s ~1901 first model, perhaps the first to subcontract
mass-produced parts for later assembly, had only 443 parts. Of course, it also looked like an open horse-carriage, with a 5-kW 4-stroke engine, 2speed transmission, and none of a modern car’s most basic passenger amenities.
625. Parts count depends on definitions: a Lotus has two main body parts, but they’re augmented by others. We normally count the elements of, say,
an underbody, door, or hood (such as foamcore, hardpoint mounting bracket, fiber, and resin) as constituting a single part. This isn’t strictly comparable to steel parts-count conventions, but broadly speaking, the approach described in Cramer & Taggart 2002 and in Lovins & Cramer 2004 reduces
body parts count by fourfold.
626. Lovins & Cramer 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
147
Implementation
Crafting an effective energy strategy: Business opportunities: Lowering risk: Manufacturing investment and variable cost
The cost of a 50,000vehicle/y greenfield
assembly plant could
drop to ~$185 million,
or $3,700/vehicle-y—
about 40% less than
GM’s most modern
plant, which in turn
is nearly one-fifth
cheaper than typical
Japanese transplant
facilities.
efficient scale of manufacturing from nearly 150,000 vehicles per year to
around 50,000. The less lumpy cashflow (plus quicker ramp-up due to
simpler manufacturing) requires and risks less investment, and the more
agile plant allows a closer and more dynamic match to customer needs.
State of the Art
vehicles’ redesign
also lowers the
minimum efficient
scale of manufacturing from nearly
150,000 vehicles
per year to around
50,000.
Investment in such
manufacturing plants
would be financially
justified, as long as
there is a market for
the vehicles.
If advanced-composite autobodies cut manufacturing capital costs to this
extent, as considerable proprietary analysis suggests they can, then the
total manufactured costs of the new generation of high-efficiency vehicles
could be competitive with today’s comparable cars.627 The composite manufacturing technologies discussed on p. 57 are already becoming available
to make carbon-composite autobody parts for less than 20% of the cost of
aerospace composites, with at least 80% of their performance.628 Fig. 20 (p.
65), for example, shows that a gasoline-engine, nonhybrid, but ultralight
advanced-composite midsize SUV could sell for just 1.6% more than the
most nearly comparable steel SUV on the 2004 market, yet use 58% less
fuel. Its Cost of Saved Energy, 15¢/gal, is one-fifth that of a conservative
estimate for the popular 2004 Prius hybrid, and yields a one-year simple
payback time at $1.50/gal.629 Fig. 20 also shows that an ultralight
advanced-composite hybrid SUV would save fuel at a cost of 56¢/gal—
still one-third below Prius’s—for a three-year simple payback. Yet these
figures are not for a midsize sedan like Prius, but for an uncompromised
midsize SUV crossover vehicle with a slightly roomier interior than a
2000 Explorer and with superior simulated crashworthiness (Box 7, p. 62).
These comparisons suggest that such an ultralight vehicle, even with
hybrid drive, could command a far broader market than Prius enjoys
today. With just hybrid drive and no ultralighting, that market is big
enough for Toyota to have slated 4% of its total 2005 production, or
300,000 vehicles, for hybrid powertrains.630 Ultralighting saves far more
fuel but costs about the same (Fig. 21, p. 66).
627. The production cost
Since a State of the Art ultralight-hybrid light vehicle made from either
analysis, described on
advanced composites or lightweight steels could thus be profitably sold
pp. 62–73 and in Technical
at a competitive cost, we conclude that investment in such manufacturing
Annex, Ch. 5, assumed
manufacturing-cost-toplants would be financially justified, as long as there is a market for the
invoice markups of 79%
vehicles. Today’s hybrids, as we just discussed, provide an encouraging
and invoice-to-MSRP
markups of 10.4%, both
hint at that market, but hybrids are still a tiny part of total vehicle sales.
consistent with industry
There are more fundamental reasons to expect robust demand for State of
margins for acceptable
the Art vehicles.
returns on capital; aboveaverage U.S. auto-assembler wages; and a greenfield plant working two shifts at 90% uptime, 64% working hours, hence 58% availability. Porsche Engineering’s analysis for the steel
industry’s ultralight-steel backstop technology (p. 67, above) is methodologically similar and, for the nonhybrid gasoline-engine ULSAB-ATV charted in Fig. 21
(p. 66), shows an extra production cost of about zero.
628. E.g., Whitfield 2004.
629. For this comparison, since there is no exact nonhybrid comparable, we assume the efficiency of a Toyota Echo and a 2004-$ marginal price of ~$1,900.
Toyota has certified $3,150 for purposes of Colorado’s hybrid tax credit (Colorado Department of Revenue, Taxpayer Service Division 2003), which is based on
marginal manufacturing cost adjusted to equivalent retail price. However, since Toyota is widely believed to have shaved its actual margin, we use the much
lower $1,900 estimate (2004 $)—close to expected MY2007 full pricing (note 178, p. 30)—to ensure conservatism. At the declared $3,150 marginal price,
Prius’s Cost of Saved Energy would be $1.35/gal (2000 $).
630. President Fujio Cho confirmed this figure, subject to production constraints, at the Traverse City CEOs’ conference, with two-fifths being non-Prius models:
Porretto 2004.
148
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business opportunities: Lowering light vehicles’ manufacturing risk
Market adoption
Both theoretical demand (estimated from decades of purchasing behavior
and recent customer surveys) and revealed behavior from the rapid initial
market entry by hybrid vehicles clearly indicate that there is a market for
higher-efficiency vehicles across the broad spectrum of light-vehicle product
lines. The size of the early-adopter segment is commonly estimated to range
from a low of 3–5%—revealed by hybrid automakers’ short-term production
plans—to a high of nearly 20% from market surveys. The market itself will
demonstrate the degree of adoption over the next five years as the hybrid
car class expands, and this will, in turn, inform automakers about potential
initial market size for broader categories of next-generation vehicles.
To the extent that the next-generation light vehicle is a truly disruptive technology that provides superior mobility services at lower cost, hence greater
value, customers can broadly adopt it with astonishing speed. A comprehensive 1999 review of disruptive technologies’ historical adoption found that
technologies using the existing infrastructure can move from 10% to 90%
capture of the capital stock within 12–15 years (p. 6, above),631 implying an
even faster increase in their share of new-vehicle sales. The new generation
of automobiles and light trucks will need new manufacturing infrastructure
if they’re made of composites (lightweight steels use conventional fabrication), but will need little concurrent investment in additional infrastructure—just routine upgrading of repair shops and skills to handle composite
bodies (more like boatbuilding than sheetmetal work) and hybrids (a little
extra test equipment and training for the electrical components). Indeed,
given hybrid vehicle manufacturers’ emphasis on making hybridization
transparent to the user,632 hybrid powertrains could in principle be adopted
with a speed comparable to, say, airbags (0–100% of the new-vehicle market
in seven years) or anti-lock braking (0 to nearly 60% in eight years).633
Since 2000, hybrids’ U.S. sales have indeed grown at an average annual
rate of 89% from a nearly zero base.634 If that kept up for seven years, it
would amount to millions of vehicles a year, but that’s not a sound analytic method. Rather, we must look to the basic principles that govern all
new technologies’ market adoption.
Innovation diffusion theory suggests that speed of diffusion and permeability of markets both rise if the new technology is perceived as superior,
easy to use, matched to customers’ needs, testable without obligation (as
one can do with rental and loaner vehicles), understandable (or unnecessary to understand—e.g. one can use a cellphone without understanding
what goes on inside it), socially compatible, low-hassle, low-risk, and
reversible (as by reselling a car). Today’s hybrid vehicles pass all these
tests nicely. We believe State of the Art cars will too. They can and should
sell because they’re better, not because they’re efficient (p. 46): not because
they’re green, but because they’re a superior product that redefines market
expectations.635
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
The same attributes
that have made other
innovations diffuse
rapidly can also apply
to superefficient
light vehicles.
State of the Art cars
can and should sell
because they’re better,
not because they’re
efficient: not because
they’re green, but
because they’re a
superior product that
redefines market
expectations.
631. Grübler, Nakićenović,
& Victor 1999.
632. Ford’s Escape hybrid
designers even included
red-light creep, which a
hybrid wouldn’t normally
have.
633. Federal Reserve Bank
of Dallas 1997.
634. R.L. Polk & Co. data
cited in Porretto 2004.
635. Most drivers don’t
know or care what their
autobody is made of, so
long as it’s safe, solid,
free of squeaks and rattles,
durable, and attractive.
Advanced composites do
all of these things better.
They also don’t rust or
fatigue, can bounce
undamaged off a 6-mph
collision that could cause
thousands of dollars’ worth
of damage to a steel car,
and can even be radarstealthy. We can easily
envisage advancedcomposite bodies attracting
not customer resistance
but bragging rights.
149
Implementation
The U.S. trucking
industry at first
appears highly fragmented. However,
the upper end of the
market is concentrated: the hundred
largest for-hire fleets
own 18% of the total
Class 8 stock.
Moreover, their
turnover represents a
whopping 55–60% of
the total demand for
new Class 8 trucks.
High-volume,
profit-hungry
new-truck-fleet
buyers can induce
truckmakers
to shift rapidly to
doubled-efficiency
models—once
the buyers
realize it’s possible.
636. This term, a combination of “fee” and “rebate,”
was probably coined by
Professor A.H. Rosenfeld at
the University of California,
Berkeley, now a California
Energy Commissioner.
The concept was suggested independently in the
1970s both by him and by
the senior author of this
report, but may have been
devised even earlier by IBM
scientist Dr. R.L. Garwin.
In Europe, feebates are
often called a “bonus/
malus” system—a term
probably due to Professor
E.U. von Weizsäcker.
637. EIA 2004; DOT 1977;
ORNL 2003.
638. ATA 2004.
150
Crafting an effective energy strategy: Business opportunities: Lowering light vehicles’ manufacturing risk
Diffusion studies also find, however, that it can take a decade or more to
achieve the first 10% market capture. The importance of the size of the
early-adopter segment thus becomes plain: if not enough people adopt
the new technology, it can languish for too many years for the automobile
manufacturers to make a profitable return on their product development
investment. This then delays even more the successive generations of
follow-on innovations.
Government policies can unblock this product evolution and greatly reduce
the risk to automakers by stimulating the market to cross the chasm of
early adoption with high speed and confidence. Starting on p. 186, we’ll
discuss and simulate the most important way to do this: policies known
as “feebates,”636 which reward a shift to more fuel-efficient vehicles without distorting choices between vehicle classes or burdening public revenues. If enacted regionally or nationally, feebates alone would create the
incentive for manufacturers to accelerate and increase their investments
in developing and making very efficient vehicles. Our policy portfolio
also includes complementary policies to increase early demand and accelerate supply of advanced technology vehicles.
We believe this combination of underlying market demand, slimmed
investment and production costs, manufacturing breakthroughs, and policy-accelerated market uptake (pp. 178–203, summarized on pp. 180–181)
suffices to create a profitable and highly competitive business model
for advanced vehicles. We therefore turn to the other two transportation
sectors clearly at competitive risk, and able to contribute notably to oil
savings: heavy trucks and aviation.
Restoring profitability in the trucking sector
Investments in fuel efficiency can help bring back financial health to the
trucking sector. Its structure and business conditions are quite favorable
for a comparatively rapid transition to more efficient fleets, because buyers and sellers of new trucks are relatively concentrated, and both have
strong economic incentives to improve fuel efficiency in order to restore
or increase profits (Technical Annex, Ch. 20).
There are two major categories of trucks: heavy trucks (Class 8, pp. 73–77)
and medium trucks (Classes 3–7, p. 77). U.S. Class 8 trucks consume ~1.5
million barrels of diesel fuel per day—10.4% of the nation’s total oil use,
and over three times the total usage of energy of Classes 3–7 (78% of whose
fuel use is in the heaviest categories, Classes 6 and 7).637 Among medium
and heavy trucks, the big ones matter most.
The U.S. trucking industry at first appears highly fragmented, with more
than 56,000 for-hire and private fleets.638 Barriers to entry are low; for example, the minimum efficient scale for LTL (less-than-truckload) operations
is 2 million ton-miles hauled annually, equivalent to just five trucks.639
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business opportunities: Restoring profitability in the trucking sector
Seventy percent of these long-haul trucks are “for-hire trucks” driven by
carriers that supply services using tractors and trailers they own.640
However, the upper end of the market is concentrated: the hundred
largest for-hire fleets own ~395,000 tractors, 18% of the total Class 8
stock.641 Moreover, these owners are the critical early adopters, since they
renew their fleets about every five years. This turnover represents a
whopping 55–60% of the total demand for new Class 8 trucks, so new
sales are highly concentrated in a small group of companies.
The trucking industry has strong incentives to save fuel, because its operations are barely profitable. The top hundred companies’ profit margins
in 1999–2000 averaged ~3.0–3.1%, and the very top performers achieved
4.9–5.0%.642 Fuel accounts for 13–22% of the costs of the typical trucking
company. The biggest are best able to buy in bulk and to equip drivers
with such sophisticated technologies as wireless systems that let them
shop for the lowest real-time price, optimizing their route for the best
fuel prices along the way as well as to minimize miles and hours traveled.
This is a business imperative, because a mere 15–20% rise in fuel costs
could wipe out the entire profit margin of the best companies. Yet not
all truckers can pass on the fuel costs to their customers, and small operators with less purchasing power often become unable to make their debt
payments. This explains the numerous trucking protests, in the U.S. and
around the world, when oil prices rise.
The converse is also true. If efficient trucks can cost-effectively reduce fuel
cost and exposure to uncontrollable fuel price volatility, then profits will
rise in proportion to the net benefit. Truck efficiency improvements across
the entire U.S. Class 8 truck stock could save 38% of its year-2000 fuel use
at an average at-the-nozzle cost of $0.25 per saved gallon of diesel fuel,
discounting the saved fuel at 5%/y real over the average life of the truck
(pp. 73–77).
Since nearly two-thirds of all new Class 8 trucks are bought by the 100
biggest companies, these companies could lower their own fuel bill by
45%643 within their average five-year fleet turnover period by purchasing
highly efficient trucks—if those were for sale.644 All adoptable measures
with marginal internal rates of return (IRR) of at least 15% would yield an
average IRR of 60%/y. 645 Using the EIA forecast of diesel fuel price in
2025—$1.33/gal (2000 $)—and the same amount of driving would raise
the average IRR to over 80%/y.
As a specific illustrative case, we estimate that if a firm like Swift Transportation Co., Inc. switched its entire fleet to State of the Art trucks, its net
profit margin would increase by ~103%. In 2003, Swift had revenue of
$2.4 billion, owned 14,344 tractors, and had a net after-tax margin of 3.3%.
By implementing all State of the Art fuel efficiency measures that each
achieve an IRR of at least 15%, Swift would save 45% of its current truck
fuel. This would bring Swift’s average new-truck mpg from 6.7 to 11.3,
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
The trucking industry
has strong incentives
to save fuel, because
its operations are
barely profitable.
639. Assuming a 25% load
factor and 67,000 mi/y of
driving. See Giordano 1997.
640. ATA 2001.
641. The hundred largest
fleets have an average
trailer-to-tractor ratio of 2.2,
and 83% have a ratio of 1.5
or greater (Transport Topics
2003).
642. BTS 2000.
643. For the specific measures discussed in the prior
section on heavy truck
technical potential and with
marginal internal rates of
return greater than 15%.
Who operates the truck at
different phases of its life
doesn’t matter if the market
that determines resale
value is reasonably efficient.
644. This figure is based on
RMI’s business case evaluation from actual 2001 data
on a sample of 100 companies, and is relative to their
fuel use in 2003. On average, these firms achieved
6.2 mpg, paid only $1.08/gal
(2000 $) net of any bulk-purchase discounts, and drove
each truck 115,000 mi/y.
645. We used data from
14 of the top 100 trucking
companies to calculate the
average price for diesel and
the average VMT (Vehicle
Miles Traveled). The 2003
average diesel price was
derived from 2001 reported
fuel costs and VMT, using
the industry average fuel
economy of 6.2 mpg and
EIA’s ratio of 2003 to 2001
diesel prices. See Technical Annex , Ch. 20.
151
Implementation
Crafting an effective energy strategy: Business opportunities: Restoring profitability in the trucking sector
The 100 biggest
companies could
lower their own fuel
bill by 45% within
their average fiveyear fleet turnover
period by purchasing
highly efficient trucks
with an average
IRR of 60%/y.
with an average IRR of 70%/y. After the fifth year, Swift’s after-tax profit
margin would rise from 3.3% to 6.7%.
646. We are also mindful
that bringing a large new
class of buyers into the
heavy-truck market, as we
propose for light vehicles,
would require major capacity expansions that could
raise short-term prices and
whose sales may not be
sustainable over the long
term as the fleet becomes
saturated with advanced
tractors.
647. Smith & Eberle 2003;
USCB 1999; KPMG LLP
2004; Navistar 2003;
PACCAR 2003.
648. Please see the
Trucking Business Case
spreadsheet (Technical
Annex, Ch. 20) for details.
152
Similarly, private carrier fleets serving companies such as Wal-Mart®,
Coca-Cola®, or SYSCO ® each account for 0.3–0.5% of the national Class 8
truck stock, and, unless the carriers pass their fuel costs on to their clients
(Wal-Mart, etc.), the private carriers also have clear incentives to upgrade
their fleets.
Most of the existing fleet, however, is owned by many smaller fleets and
independent truckers, who tend to buy their trucks used from the large
private carriers and run them for the rest of their useful life. Owning the
trucks far longer than private carriers, these smaller operators drive twice
as many cumulative total miles per truck; yet they’re also unlikely to have
the cash or credit to upgrade their vehicles early. These truckers, responsible for ~60–65% of total Class 8 fuel use, will have to wait for State of the
Art vehicles to trickle down to the used-truck market. Nevertheless, the
five-year resale cycle is much quicker than for light vehicles or airplanes,
so although policies could readily be devised to accelerate adoption by
bringing used-truck buyers into the new-truck market, we have chosen
not to do so.646 Instead, we focus on the ~23% of the Class 8 operators that
buy ~60% to ~75% (averaging nearly two-thirds) of all new trucks and
are eager customers for efficiency improvements. Given the very positive
customer economics, we expect that the first manufacturer to make this
shift will gain substantial market share.
The Class 8 truck manufacturing market is highly concentrated: four major
players (DaimlerChrysler, Volvo, Navistar, and PACCAR) sell 99.5% of
U.S. heavy trucks. Unlike cars, the efficiency improvements described
on pp. 73–77 do not require a major change of platform design nor new
product lines. Instead, the improvements are about equally divided
between lightweighting, making the body more aerodynamic, and modernizing the engine.
While we don’t have firm capital estimates of the investment cost, clearly it is
comparatively modest. For the four big truckmakers, we estimate that a
cumulative total investment of ~$18 billion would be required during
2005–25 to retool 15–24 plants, or ~$750 million per plant647—about a halfyear’s worth of revenue per plant. This proposed investment is well within
the range of what these manufacturers already plan to invest. When annualized, it is equivalent to 50–70% of current (2003 actual) and forecast (2004–06)
investment rates that those four OEMs already specifically target at high-performance vehicles and next-generation diesel engine technologies, often using a
more incremental approach. In addition to these costs, truckmakers would
incur a ~$3–5 billion total charge for product-line development costs.648
RMI expects that an informed market will make these changes on its own.
The major customers for new trucks receive high returns from more effi-
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business opportunities: Restoring profitability in the trucking sector
Implementation
cient vehicles, as do the truck manufacturers from increased profits due to
selling more expensive trucks. The technology improvements necessary to
make the vehicles more aerodynamic and lightweight are substantially
easier technically than for cars (Technical Annex, Ch. 6), and most are well
proven. These innovations account for two-thirds of trucks’ potential fuel
savings. The other third comes from engine improvements, which will
require more R&D, but that’s the core product-development business of
such capable firms as Cummins, Caterpillar, and Detroit Diesel.
Our judgment is that the main missing ingredient for radically increased
efficiency in the trucking sector is better customer information on what’s
possible. RMI’s recent conversations with the heads of two very large
Class 8 fleet operators substantiate this hypothesis. These savvy business
leaders were astonished that more than a few percent of truck energy
could be profitably saved. On learning of the State of the Art potential, their
basic reaction was, “Let’s build one, and if it works, we’ll simply tell the
truckmakers that we want them to make such trucks for us.” When our
consulting practice previously advised one of these same firms on an
experimental building, its design required a certain technology that could
readily be made but hadn’t been brought to market. When we asked the
leading vendor’s sales department for one, they said, “Sorry, sir, it’s not in
the catalogue.” When we replied, “Our client is X Corporation, and if they
like this product, they’ll buy a truckload a day indefinitely,” the answer
instantly changed to “Yes, sir! When do you want it?” That energy-saving,
profit-enhancing product was duly delivered, successfully tested, and
widely propagated.
The opportunity to do the same with Class 8 trucks is starting to be validated. In 2004, one of the most innovative, engineering-intensive, and
successful bulk carriers in the U.S.648a redesigned a Class 8 tractor-trailer
combination from scratch in collaboration with a major truckmaker. The
goals—25% higher payload, doubled fuel economy, enhanced safety and
driver ergonomics, and reduced emissions—did turn out to appear feasible, and a prototype is planned to be built in late 2004. Interestingly, the
technology suite didn’t include the Cd reductions and superefficient
engines emphasized on pp. 73–77 above, but is expected to achieve comparable results by emphasizing careful integration of off-the-shelf
improvements. This implies that our State of the Art portfolio conservatively omitted some collectively significant technological opportunities.
The business opportunity for the truck manufacturers is not selling more
trucks in the U.S. per se, but selling trucks that provide higher value to
their customers, yielding decisive advantage in margin or market share.
The same product improvements will also make the trucks more competitive in other countries where the diesel prices are higher, notably in
Europe, where the reduced CO 2 and other emissions would also provide
a strong marketing advantage. For medium trucks, major development
efforts are already underway in Europe and Japan to reduce radically the
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
RMI expects that an
informed market will
make these changes
on its own.
The main missing
ingredient for
radically increased
efficiency in the
trucking sector is
better customer
information on what’s
possible.
648a. Logistics Management, Inc. (Bridgeview, IL).
Tom Wieranga, President
(contactable via RMI),
personal communication,
16 August 2004.
153
Implementation
Next-generation
airplanes save fuel
cheaply (or even
better than free),
but legacy airlines
can’t afford them.
Federal loan guarantees can make such
acquisitions financeable, and should
be coupled with
scrappage of the
inefficient old planes
now parked.
Fuel savings from
new airplanes are
very cheap per gallon, but new planes
are expensive. The
legacy airlines are
too broke to buy many
new planes, so when
the profitable discount carriers buy
very efficient new
planes, they displace
their high-cost competitors even faster.
649. U.S. legacy airlines’
total cost is 9.5–11.5¢/seatmile; low–cost carriers’,
6–8¢/seat-mile.
650. M.W. Walsh 2004.
651. New York Harbor jet
fuel prices were used (EIA,
2004e, p. 26, Table 15).
652. Tully 2004, p. 101.
653. Financial Times 2004.
154
Crafting an effective energy strategy: Business opportunities: Restoring profitability in the trucking sector
noise and emissions of urban delivery trucks; this is accelerating global
demand for the most advanced technologies, such as fuel cells. And of
course the same spectrum of demands occurs in the Pentagon’s truck
fleet, perhaps the world’s largest (p. 88, note 430), with the added incentive of the huge logistics costs for delivering fuel. When one also considers DoD’s need for superefficient diesel engines for medium and heavy
tactical vehicles, using military procurement to insert advanced technologies rapidly into the heavy-truck market faster becomes a key enabler for
military transformation.
Revitalizing the airline and airplane industries
Fuel costs are the airline industry’s biggest variable cost except labor.
Because of their high volatility, fuel costs are the industry’s financial
Achilles’ heel. Per passenger-mile, you pay airlines about the same as you
pay to own and run your car, but like low-income households, airlines
gush red ink whenever fuel prices spike up. Meanwhile, overcapacity
is depressing the near-term growth prospects of airplane manufacturers.
How do we address both problems, so as to revitalize the airline industry
and reduce oil dependence at the same time?
The key to this puzzle is that fuel savings from new airplanes are very
cheap per gallon, but new planes are expensive. The legacy airlines are
too broke to buy many new planes, so when the profitable discount carriers buy very efficient new planes, they displace their high-cost competitors649 even faster, accelerating the market exit of the legacy carriers and
the cost savings to customers.
If policymakers decide that they wish to save the legacy airlines, a good
way to give them a better chance of survival—better for society than, say,
simply throwing federal dollars at their operating deficits or assuming
their $31 billion in pension obligations650—would be an innovative loan
guarantee program for the airline industry to purchase or lease new fuelefficient aircraft (Box 16). Current-dollar jet fuel prices rose 54% to
$1.18/gallon between May 2003 and May 2004, increasing fuel costs to
2.13¢/seat-mile.651 As we noted on p. 17, each time the jet fuel price
increases 1 cent, it costs U.S. airlines $180 million a year.652 Higher fuel
prices cost the three largest U.S. carriers—United, American, and
Continental—an additional $700 million in 2004.653 But if new efficient aircraft used Conventional Wisdom technologies, U.S. airlines’ fuel cost would
drop 29% to 1.51¢/seat-mile, saving $6.2 billion per year. Those savings
would rise to $8.4b/y with State of the Art planes, which would cut fuel
costs to 1.14¢/seat-mile, even at mid-2004 fuel prices.
The economics of fuel efficiency are compelling. Over its 30-year life, one
7E7 will save over $7 million in present-valued fuel costs compared with
the 767 it replaces (or $5.3 million vs. the A330 with which it competes),
yet its capital cost is comparable or lower.654 If United had reconfigured
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Crafting an effective energy strategy: Business opportunities: Revitalizing the airline and airplane industries
just 9% of its 2003 fleet with similarly efficient planes, its net income
would have increased by $34 million.655 (State of the Art planes could
roughly double those savings.) Even incremental retrofits can be attractive. For example, Southwest is executing a plan to attach blended
winglets to all of its 737-700s at a cost around $710,000 per plane.656
The project will improve each plane’s fuel efficiency by 3–4%, saving
~$113,000 more than the winglets cost (present-valued at 5%/y real over
30 years). If Southwest applied this technology to its entire 2003 fleet of
388 aircraft, it would invest nearly $300 million to cut operating costs by
0.6%, paying back in 8 years and earning a real IRR of ~12%/y.657
So if fuel savings are profitable, why aren’t all airlines updating their
fleets like Southwest, especially with new planes?
The answer lies in the structural changes in the U.S. airline industry and
the financial state of the U.S. airlines. The major U.S. carriers are facing
devastating competition from discount carriers whose lower-cost business
model is based on a more efficient point-to-point service, lower embedded labor costs, and lower non-labor operating costs. Low-cost carriers’
market share has risen from 5% in 1987 to 25% in 2004. The market capitalization of the six biggest discounters is now $18 billion, far higher than
the $4 billion market capitalization of the six largest traditional airlines.658
Southwest Airlines alone has a tenth the market share and a twelfth the
revenue of the Big Six combined—but nearly three times their combined
market capitalization.
The major airlines, having borrowed ~$100 billion to stay alive,659 are now
carrying so much debt that either they’re in bankruptcy or their credit
ratings have slipped to junk-bond status. This makes it hard for them to
finance the new fuel-efficient (and often more lightly crewed) jets they
need. And if they were to do so, the present-valued lifetime fuel savings
are only ~6% of the capital cost of a new plane—critical to operating
margins, but hardly enough by itself to justify turning over the fleet when
cash is scarce.660
If new efficient
aircraft used
Conventional Wisdom
technologies,
U.S. airlines’ fuel cost
would drop 29%,
saving $6.2 billion per
year. Those savings
would rise to $8.4b/y
with State of the Art
planes.
Southwest Airlines
has a tenth the market
share and a twelfth the
revenue of the Big Six
combined—but nearly
three times their
market capitalization.
654. The 7E7 replaces the
current 767. Even at the EIA
estimate for 2025 jet fuel
prices of 81¢/gal (2000 $,
based on $26/bbl crude),
its lifetime fuel savings at
the U.S. airlines’ current
weighted-average cost of
capital (13%/y) are $7.4 million, based on flying 900
flights/y, each of 3,500 nautical miles. From airlines’
perspective, the fuel savings are only 6% of the
~$120 million price of a
new 7E7. We assumed a
nominal 30-y aircraft life
(Ferguson 2004).
655. This illustrative example assumes that United Airlines replaced its 37 767-300s and 10 767-200s with 47 7E7s. At a fuel saving of 23.8% vs. a 767-200 and
17.3% vs. a 767-300 (M. Mirza, Boeing Economic Analysis Dept., personal communication, 11 June 2004, based on a standardized 2,000-nautical-mile trip at
the 2000 average load factor of 0.72), 7E7 would have saved 18.7% of the fuel used by those 47 aircraft (8.8% of United’s fleet). Assuming an even distribution
of miles flown across United’s 532-aircraft fleet, this replacement would have saved 18.7% of $183 million/y (8.8% × $2,027 million/y fuel expense), or $34.2
million/y. (Fuel expense and fleet data from UAL Corporation 10-K, filed 2 March 2004.)
656. Martin, Rogers, & Simkins 2004. Our IRR calculation assumes Southwest’s weighted-average real cost of capital is 10%/y,
close to its actual WACC of 10.5% at mid-2004, and assumes a constant real fuel price of 81¢/gal (2000 $), equaling EIA’s 2025 forecast.
657. Calculations based on the expected saving of 110,000 gal/plane-y at the EIA 2025 price forecast of 81¢/gal.
658. Tully 2004.
659. Jenkins 2004.
660. Of course, it’s not worth buying a new car just for its fuel savings either, but when one needs a new car anyway, its potential fuel savings are a much
larger fraction of total capital cost than for a new airplane. Airplanes are usually more fuel-efficient than cars, counting actual load factors for both (see the
right-hand seat-mpg axis of Fig. 25 on p. 81; actually planes have about three times the load factor of cars). However, their capital cost per seat is two orders
of magnitude higher, because they travel ten times faster, for 2–3 times the lifetime and ~20–30 times the operating hours, and are made at five orders of
magnitude lower volume.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
155
Implementation
Efficient new planes
like the 7E7 and its
successors could
easily be stalled by
the least efficient
one-fifth of the fleet
that’s now parked.
If traffic picks up and
those parked planes
resume service,
they’ll not only waste
fuel and continue to
bleed operating
budgets they’ll also
slow the adoption
of far more efficient
new models, and
hence the development of next-generation State of the Art
airplanes too.
That’s why we propose to link the
scrappage of inefficient parked planes
to loan guarantees
for financing efficient
new planes.
Crafting an effective energy strategy: Business opportunities: Revitalizing the airline and airplane industries
While major airlines are cutting back on plane orders, the low-cost carriers
have orders for 500 new jets over the next 5 years—more than the 357 jets
Continental Airlines currently flies.661 The low-cost carriers would be well
advised to consider very fuel-efficient planes too. As discount carriers
move from regional point-to-point service (whose short stages increase
fuel use per seat-mile) to transcontinental flights, their flights surpass
the trip lengths at which the value of fuel-efficient new planes starts to
increase more and more rapidly. Efficiency saves more on longer trip
lengths because the base benefit of more efficient planes is compounded
by the reduced amount of fuel they must carry for the longer trips vs. their
heavier and less efficient counterparts. On a standard 2000-mile flight, a
typical existing widebody would have to carry at least 11,594 gallons of
total fuel, of which 818 gallons are needed just to carry fuel. This is only
a 7% overhead, but for longer missions, the fuel needed to carry fuel rises
16: Flying high: fuel savings arbitrage
The airlines are starting to draw on
$10 billion in federal loan guarantees
to help them out of bankruptcy.
While the airlines have used bankruptcy to address labor costs, they
have not fixed the structural problem
of fuel cost. The economics of fuel
efficiency are compelling, but the
major airlines lack the balance-sheet
strength to retrofit their fleet. Rather
than simply bail out the airlines, and
hope they recover, why not double
down on the bet, and finance the
restructuring of their fleet to achieve
far greater national benefits?
Our proposed program calls on commercial banks and the federal government to arbitrage discount rates
and liquidity by providing financing
for the airlines to restructure their
fleet toward increased fuel efficiency. Airlines could receive federal
loan guarantees to buy or lease any
new aircraft that meets a high target
for fuel efficiency, from any manu-
facturer (competition is good), provided that for each plane so
financed, an inefficient parked plane
is scrapped. (A trading system would
allow buyers who don’t own parked
planes to satisfy the scrappage obligation.) Fuel savings would help
repay the commercial financing, just
as energy efficiency financing does
in buildings and factories. This program addresses the overcapacity
problem that is stifling the sale of
new aircraft, and would stimulate
orders for new planes.
Airlines would be allowed either to
purchase or to lease the new fuelefficient aircraft, but would be
encouraged to enter into operating
leases, thereby reducing their onbalance-sheet debt and improving
their ability to borrow. A program of
this nature need not be limited to the
legacy airlines; a similar program
could be crafted for the new planes
bought by the low-cost carriers.
661. McCartney 2004.
156
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business opportunities: Revitalizing the airline and airplane industries
Implementation
until at the limit of a 747’s globe-girdling range, nearly two gallons must
be loaded to have one gallon left at the destination. By contrast, a State of
Art widebody plane would need to carry only 6,716 total gallons of fuel
while transporting 173 more passengers.662
If the next generation of planes can deliver fuel efficiency at below the EIA
projected 2025 jet fuel price of 81¢/gallon, as is clearly the case (pp. 79–83),
then the discounters should spend the extra money to acquire these planes.
If they want to stay competitive, the future is already here: a 7E7 can save
one-fifth of the fuel used by the 767 it replaces, but at the same or lower
real capital cost.663
The parked
airplane fleet is
worth more to society
dead
than alive.
Yet rapid adoption of efficient new planes like the 7E7 and its successors
could easily be stalled by the least efficient one-fifth of the fleet that’s now
parked. If traffic picks up and those parked planes resume service, they’ll
not only waste fuel and continue to bleed operating budgets (making airlines even less able to escape from the fuel-cost trap); they’ll also slow the
adoption of far more efficient new models, and hence the development of
next-generation State of the Art airplanes too. That’s why, in Box 16, we
propose to link the scrappage of inefficient parked planes to loan guarantees for financing efficient new planes. The parked fleet is worth more to
society dead than alive—counting not just their fuel waste, but the opportunity cost of their delaying the adoption and development of ever more
efficient successors—and this linkage seems a simple way to activate
bounty-hunters.
Getting generation-after-next planes off the ground
The underlying economics of State of the Art blended-wing-body aircraft are
compelling even when compared to excellent Conventional Wisdom planes
like the 7E7 they’ll ultimately replace. State of the Art planes will save 30%
more fuel than Conventional Wisdom planes, delivering fuel savings at only
43¢ per gallon of saved jet fuel, far below the EIA price of 81¢ per gallon in
2025. We therefore expect that these planes could be adopted by the airlines. The question is when, and how that schedule can be accelerated.664
Even more efficient
successor planes
should be readied
promptly to start
saving even more fuel
after 2015. Military
requirements can
accelerate their
development.
By 2025, EIA projects a 60% increase in passenger miles and a 130%
increase in cargo miles. To meet this increased travel demand, EIA projects about 15,000 new airplanes to be sold between 2005 and 2025, a third
662. Calculated by using average-widebody and SOA aircraft gal/seat-mile data from Technical Annex, Ch. 12. Seating (295
passengers) and incremental fuel burn (0.005 gal per flight-hour per pound of weight added) are determined based on a
widebody aircraft fleet made up of 31% 747-400s, 51% 767s, and 18% 777s. Fuel reserves are based on Federal Aviation
Requirement 91.151, which mandates 30 minutes of cruising speed fuel for daytime flights.
663. The weighted-average price of all 767-200ER, -300ER, and -400ER airplanes placed in world service in the past five
years was $119 million (2000 $), while the 7E7 is estimated to sell for $112±4.7 million (2000 $), implying a negative Cost of
Saved Energy.
664. Packaging issues may also arise in small sizes, such as for regional aircraft, but the sort of advanced-composite construction method used in the Lockheed-Martin Skunk Works’ advanced tactical fighter in the mid-1990s (p. 62, note 326)
should help to fit more passengers and cargo into a smaller, lighter, and cheaper blended-wing-body airplane.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
157
Implementation
Crafting an effective energy strategy: Business opportunities: Revitalizing the airline and airplane industries
The overarching
national aviation goal
should be to ensure
that new planes
bought in the short
term are as efficient
as they can be (like
7E7); that State of the
Art successor models
are developed and
brought to market
with due deliberate
speed; and that both
airframe manufacturers and their customers find it advantageous and feasible
to adopt the most
efficient airplanes
available at any
given time.
of which will be regional jets.665 Airlines are willing to pay more for reductions in operating costs: the investment per seat for long-range aircraft
increased 130% during 1959–95.666 But some airlines need financing, and
all need superefficient planes to choose from.
Worldwide commercial airline production is largely a duopoly comprising
Boeing and Airbus. These companies are in a continuous high-stakes
engineering design competition to beat the rival’s performance. Currently
they each have a new plane in development for release in the next four
years (Box 14). Airbus is set to release the 555-seat A380, a plane designed
to compete with Boeing’s 747, around 2006. Boeing in turn is designing the
7E7 Dreamliner to replace its own 767 and 757 models and compete against
Airbus’s A330. Airbus is worried enough to begin to design a more-fuelefficient engine upgrade for its A330, though that platform’s higher weight
and drag will limit its ability to compete with 7E7.667 Both manufacturers
chose to target the competition’s signature aircraft, and each claims to
have surpassed the other’s current operating efficiency by at least 15%.
Development costs are around $10–12 billion for a new model aircraft.668
Airbus projects that the A380 project will not reach breakeven until the
251st plane is sold (even longer if list prices are discounted). Based on the
adoption pattern of the 747-400 released in 1989, this volume may take
longer than four years to achieve.
Technology innovation is not just about improved efficiency, however.
Technological innovation enables, and sometimes creates, new business
models. Airbus is betting on the continuation of the existing hub-andspokes business model by even larger fleets, justifying gigantic airplanes.
Boeing is betting that the discounters’ point-to-point regional business
model will spread to transcontinental and intra-regional flying. Boeing
may also be contemplating design variants larger or smaller than the 7E7
base model to hedge its market-structure bet, whereas the A380 will offer
less size flexibility. At any size, Boeing’s more efficient technology should
be attractive to airlines facing volatile and possibly high fuel prices.
665. EIA 2004.
666. Babikian et al. 2000.
667. Lunsford & Michaels
2004.
668. Wallace 2003; BBC
News 2000; note 609.
669. Flug-Revue Online 2000.
158
RMI expects that this battle will play itself out over the next five to ten
years. If Boeing begins to gain the advantage, then its business strategy
will have been proven by the market. We expect this to spur investment
by all airline manufacturers in a more efficient generation of aircraft.
A complete cycle of new airplane development takes up to about a
decade, followed by another couple of decades to replace older models.669
Therefore, State of the Art planes would not come on line until 2015 or
beyond, regardless of technological and manufacturing improvements,
and much of their benefit comes after 2025. Our scenario analysis below
assumes that after 2015, new State of the Art planes would be phased into
the fleet. Of course, feebate-like incentives could accelerate this adoption,
but may not be necessary. The overarching national aviation goal should
be to ensure that new planes bought in the short term are as efficient
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Business opportunities: Revitalizing the airline and airplane industries
as they can be (like 7E7); that State of the Art successor models are developed and brought to market with due deliberate speed; and that both
airframe manufacturers and their customers find it advantageous and
feasible to adopt the most efficient airplanes available at any given time.
An obvious way to accelerate State of the Art civilian airplane development
is smart military procurement, since blended-wing-body or “flying wing”
designs—under development by Boeing, NASA, and others since the early
1990s—and their associated technologies, such as next-generation engines,
also have broad military applications, initially for tankers, weapons systems, and command/control, and later for heavy lift and cargo applications. The blended-wing-body concept is under continuing study by
Boeing’s Phantom Works and Integrated Defense Systems branches for
military use. A 17-ft-wingspan scale model was flight-tested in 1997 in
collaboration with Stanford University, and an improved 21-ft-wingspan
model is under development with Cranfield Aerospace (UK). DARPA and
other agencies are helping with concept development for military applications, and Boeing and its competitors would happily build on that expertise to bring medium-to-large blended-wing-body airplanes to the commercial market as market conditions warrant. Thus with airplanes as with
heavy trucks, the key enabling technologies are of such strong military as
well as civilian interest for both cutting costs and transforming capabilities
(pp. 84–93) that military science and technology development should be
one of the leading elements of any coherent national effort to displace oil.
Across the range of land, sea, and air platforms, this is most true for
advanced lightweight materials, as we see next.
Creating a new high-technology industrial cluster
Technological advances are generally and rightly considered the main
engine of economic growth.670 State of the Art ultralight-hybrid vehicles
and their associated advanced technologies are in the same tradition as
the technological changes that were so vital to the U.S. economy in the
twentieth century. Throughout their lifecycle, these vehicles will consistently favor high-productivity production processes and supply highskill, high-wage, high-value-added jobs with large and widely distributed
economic multipliers.
Implementation
With airplanes as
with heavy trucks,
the key enabling
technologies are of
such strong military
as well as civilian
interest for both cutting costs and transforming capabilities
that military science
and technology
development should
be one of the leading
elements of any
coherent national
effort to displace oil.
Advanced ultralight
materials, engines,
and other efficiency
technologies are
the foundation of
oil savings and a
stronger industrial
and employment
base. Their accelerated military and
civilian development
should be vigorous,
coordinated,
and immediate.
Automaking has undergone several major transformations before, often
triggered by new materials.671 Our proposal for the transformation of the
transportation sector is underpinned by technological improvements in
670. For example, see Anton, Silbergilt, & Schneider 2001; see also Christensen & Raynor 2003, which argues that corporations accrue higher market value
from their ability to adopt innovations.
671. Amendola 1990: “The choice of materials has always been one of the major technical problems in planning, designing and manufacturing a car.…
[T]he very history of the automobile industry is rich in innovations strictly related to decisive choices about materials used. For instance, the introduction of
the Model T Ford in 1908, which is often referred to as the beginning of the modern automobile industry, was associated with a very important innovation in
the area of materials: use of a high-strength vanadium steel alloy in critical chassis components. This innovation is considered by historians as ‘the fundamental chassis design choice.’”
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
159
Implementation
Crafting an effective energy strategy: Creating a new high-technology industrial cluster
The military and
aerospace sectors
are the most likely
candidates to build
the initial primary
market demand that
would enable the
advanced materials
sector to gain
the requisite scale
economies.
advanced materials and their manufacturing techniques, powertrains
(especially using electric traction), power electronics, microelectronics,
software, aerodynamics, tires (another materials-dominated field), and
systems integration. Collectively, these form a new high-technology
industrial cluster that will expand U.S. competitiveness beyond the transportation sector alone—much as the development of the microchip, crossfertilized with other technologies, has created the largest and most
dynamic sector of the modern economy, and the information revolution in
turn has transformed the entire economy.672 How can another such co-evolution (in the automotive sector and beyond) be encouraged in order to
synchronize the industrial development strategy? And rather than stifling
this evolution with planning, how can creative policy maximize opportunities for the broadest and most durable kind of wealth creation?
RMI’s analysis strongly suggests that the military and aerospace sectors
are the most likely candidates to build the initial primary market demand
that would enable the advanced materials sector to gain the requisite scale
economies. The diffusion of military innovations into wide civilian use has
encouraging precedents. The best-known example is the microchip:673
672. This topic is discussed
by Davis, Hirschl, & Stack
1997. In a July 2001 interview, Davis states that
“…the way that we make
things—the tools and the
technology and the science
and the technique—all
those things are such a
fundamental part of the
economy that when they
change everything else
changes with it. The main
player today is the microchip, which was commercially introduced around
1971, because it makes so
many other breakthroughs
in human activity possible—
from manufacturing to agriculture, to medicine, transportation, how we produce
all forms of culture.”
673. NAS/NRC/CETS 1999,
p. 137. In FY1977, DoD R&D
represented 40% of
federal spending for basic
research and over 70% of
all federal investment in
microelectronics and electrical engineering (NAS/
NRC/CETS 1999, p. 138).
In 1976, U.S. military purchases accounted for 17% of IC [integrated-circuit] sales
worldwide ($700 million out of total sales of $4.2 billion)—a significant market
share that gave DoD leverage in defining product specifications and directions.
In the next 20 years, the U.S. military market increased only marginally, to $1.1
billion, while the commercial market exploded to $160 billion. The military market now [in 1999] accounts for less than 1% of sales, and the commercial market
has become the dominant force in setting IC product directions. Although lower
prices have resulted, the DoD is now compelled to use commercial IC products
and adapt them to meet military requirements, as necessary.
Such investments often have surprising and serendipitous spin-offs.
Military R&D in advanced aero-engine turbines is obviously the basis of
today’s modern commercial aircraft, which could hardly get off the
ground without those lightweight, fuel-efficient, high-bypass engines.
Less obviously, those commercial engine technologies in turn are the basis
for the combined-cycle gas-fired power plants that have rapidly transformed the global electric power industry.
The advanced materials we envisage for efficient cars and trucks, especially the carbon-fiber composites, are historically based on aerospace technology. The business challenge arises because aerospace applications typically
have about a thousand times smaller volume and a thousand times higher
cost than automotive ones. Direct technology transfer is therefore insufficient; R&D is needed for mass production of advanced-composite structures that meet the automotive industry’s requirements of high volume
and low cost. As noted in our earlier discussion of the lightweighting revolution (p. 57), such efforts in the private sector already show promise, and
military attention could greatly accelerate their application.674
674. Whitfield 2004.
160
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Creating a new high-technology industrial cluster
Non-aerospace military needs, especially for lightweight land and sea platforms, are an important pathway to this commercialization because they
often need higher production volumes and lower costs than military aircraft. As previously discussed (pp. 84–93), military mission requirements
focus strongly on lighter, stronger, cheaper, and more energy-efficient vehicles to fit today’s rapid-response and agility-based doctrine and to lower
logistics cost and vulnerability. The military has the R&D budget to support the development of advanced materials (composites and lightweight
steel), and already does so to a degree. The military clearly has the scale.
If DoD were to adopt the changes proposed to its Humvee (HMMWV)
production alone, a 10,000-unit yearly production run of the multipurpose
vehicle would require 5 million pounds of carbon fiber. This represents
over 6% of the current worldwide production capacity of carbon fiber for
all applications (80 million pounds). By 2025, the U.S. automobile industry
alone could demand ~1.7 billion pounds of carbon fiber, 21 times current
worldwide production, corresponding to compound growth of 15%/y.675
And this rapid expansion is feasible: composite industry sources estimate
that within five years the industry could be ready for cost-effective mass
production of carbon-fiber-based civilian and military vehicles.676
Implementation
By 2025, the U.S.
automobile industry
alone could demand
~1.7 billion pounds
of carbon fiber,
21 times current
worldwide production, corresponding
to compound growth
of 15%/y. The industry
could be ready for
cost-effective mass
production of carbonfiber-based civilian
and military.
The synchronized timing of the co-evolution is important. For example,
Ford’s first round of Escape hybrid SUVs is limited not by the market or its
manufacturing capacity, but by Sanyo’s nickel-metal-hydride buffer-battery
manufacturing capacity, because soaring demand for Japanese hybrids
keeps the supply chain rather fully occupied.677 Therefore, military technology support should be launched now in order to have the capacity ready
in time for the automakers’ shift. To minimize exposure to the cumbersome
DoD budget process, early commitments should focus on DARPA and the
more agile R&D and early-application Service organizations.
The new industrial cluster would bring national benefits far beyond military prowess and budget savings. It would create a significant number of
high-technology manufacturing jobs, which we estimate to be roughly
analogous to the labor intensity of the chemicals sector (2.3 jobs per million
dollars of annual revenue using the narrowest definition and excluding all
multipliers). The 2025 carbon-fiber demand mentioned above would fetch
~$8 billion per year, creating ~20,000 new direct jobs.678 For our projected
State of the Art vehicle production volume, these jobs may either go to the
steel sector for new lightweight steel or to the polymer composite sectors,
or to some mixture; market competition will sort that out, but either way,
675. Based on 8 million SOA cars with an average of 212 lb of carbon fiber per car—196 lb/car (Hypercar, Inc. proprietary mass-budget analysis, D.R. Cramer,
personal communication, 11 May 2004) times 1.08 scaling factor (p. 96, note 164, above).
676. Levin 2002. However, big speculative price swings that now deter investment in carbon-fiber production capacity would need to be smoothed by making
futures and options markets in structural carbon.
677. MSNBC News 2004.
678. Calculated on the basis of 8 million State of the Art cars built in 2025, each with 212 lb of carbon fiber, and one manufacturing job created for every
43.4 tons of carbon fiber produced. Job creation numbers derived from Alliance of Automobile Manufacturers economic contribution study on plastics and
rubber producers (McAlinden, Hill, & Swiecki 2003, p.25).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
161
Implementation
Crafting an effective energy strategy: Creating a new high-technology industrial cluster
it means good jobs.679 There are probably many times more jobs, too, in
converting raw carbon fiber into cars than in making the fiber. Looking
forward to the hydrogen economy, which ultralight vehicles can help accelerate (pp. 233–234), several studies on fuel-cell manufacturing predict on
the order of hundreds of thousands of new jobs.680 Clearly, this cluster of
new automotive-and-energy-related technologies could be an important
engine of economic growth. It is indeed plausible that jobs lost in, say,
petroleum refining and petrochemicals would be more than offset by new
jobs that apply broadly similar skills to polymer manufacturing and application—especially as mass-produced fuel cells, too, switch their materials to
molded and roll-to-roll polymers.
Rural and smalltown America can
gain enormously in
income, jobs, and
stability through biofuel production and
related revenues—
while the country
gains half a
Saudi Arabia’s worth
of stable, uninterruptible, all-domestic
fuel supplies.
679. For a discussion of
job impacts from carbon
composites, see Lovins et
al. 1996, Ch. 6.
The national benefits that will grow out of the new advanced-materials
industrial cluster go far beyond the direct jobs it creates. These new technologies have the potential to be as pervasive and transformative as
plastics were in the 1960s. Advanced polymer composites have already
entered boatmaking, military and civilian airplanes, bicycles, and sporting goods, and are now poised to enter the automotive sector and beyond.
Lightweight steel also has the potential to extend steel usage well
beyond the realm of traditional applications. Whichever advanced light
material comes to dominate future automobile construction (most likely a
diverse mix, each used to do what it does best), the materials applications
throughout society will extend far beyond their originally intended
mobility applications.
Restoring farming, ranching, and forest economies
Farming has never been an easy business, and the decline in U.S. agricultural communities’ population, income, and cultural vitality has been
relentless. An average of 450,000 agricultural jobs (farm proprietors and
employees) have been lost each decade since 1970.681 The U.S. and many
other OECD countries have responded to this decline with ever-increasing subsidies for the agricultural sector. These subsidies not only distort
markets and farming practices,682 but also hobble the global development
process, depressing poor countries’ food exports and domestic-market
food production to a degree that recently triggered a revolt and blocked
further trade liberalization.683 Traditional thinking on U.S. biofuels is that
680. “PricewaterhouseCoopers predicts that by 2013 the North American fuel cell industry will represent 108,000 direct and indirect jobs associated with
manufacturing stationary fuel cell units….” (Fuel Cell News 2003). The Breakthrough Technologies Institute predicts 189,000 direct and indirect jobs
produced by the fuel cell industry (BTI 2004).
681. Calculated from Bureau of Economic Analysis, Regional Economic Accounts, Local Area Personal Income (BEA, undated).
682. The U.S. currently provides a $0.53/gallon ethanol subsidy, and proposed a $1/gallon biodiesel subsidy in the 2002 Senate Energy Bill (Peckham 2002).
Europe has used a combination of oilseed-production subsidies and biodiesel tax exemptions (p.106) to ensure competitive pricing and therefore, market
adoption of biodiesel (CRFA, undated). However, “...tying government subsidies to commodity prices has distorted the agriculture market and made farmers
dependant on government handouts” (Hylden 2003).
683. The 1 August 2004 framework agreement to remove agricultural export subsidies (which the U.S. denies it has) and substantially reduce direct crop
subsidies (which have powerful political defenders) will be slow and hard to convert into actual desubsidization in the U.S. and EU. However, the agreement
increases the need for a different and trade-equitable way to strengthen rural economies.
162
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting an effective energy strategy: Restoring farming, ranching, and forest economies
Implementation
they are yet another in the long line of subsidized crops, representing
an agricultural bailout by urban and suburban dwellers.684 But this view,
based on corn-ethanol conversion, is now badly outdated (p. 103).
Once cellulosic biofuel technology lowers the cost of ethanol below the
equivalent crude oil price benchmark of $26/bbl, as has already occurred
for sugarcane ethanol in Brazil (p. 105), the game changes fundamentally.
It’s no longer about who can lobby best for crop subsidies, but about
strategic investment in a new and more secure rural-based domestic fuels
infrastructure that competes, without subsidy, even with today’s subsidized gasoline (i.e., producible for $0.75/gal gasoline-equivalent at the
plant gate, p. 103). The benefits of such substitution are well established:
greater energy security, increased agricultural employment, and reduced
carbon intensity among others.685 The business question is whether or not
next-generation biofuels and biomaterials make financial sense to farmers
and agricultural processors.
From the farmer’s perspective, the fundamental business decision is
based on the value per acre vs. alternative agricultural crops. Promising
crops proposed for biofuels (p. 107) include switchgrass and such shortrotation woody crops as hybrid poplar and willow. Based on our State of
the Art analysis, a cellulosic ethanol cost of $0.61/gallon ($0.75/gallon of
gasoline equivalent) implies that biofuel refiners can pay $54 per dry ton
for these crops and still make an adequate return on capital from their
operations. A representative switchgrass yield of 7–10 ton/acre-y yields
revenue of $431–637/acre-y.
That revenue from the biofuel crop, much or even most of it from land
otherwise unsuitable for or reserved from conventional cropping, is just
the beginning. Much of the same land can be simultaneously used for
windpower in the nation’s extensive windy areas, notably on the High
Plains. Farmers can also capture three kinds of carbon credits: from the
fossil fuel displaced by both the biofuels and the windpower, and from
the net increase in soil carbon storage. Soil carbon accumulates because
biofuel-feedstock farming using USDA-recommended soil conservation
practices can be better than carbon-neutral: such farming methods can fix
airborne carbon (CO2) into soil to improve fertility, water retention, and
biodiversity (which in turn can substitute for fertilizers, pesticides, and
other costly inputs). USDA’s new Conservation Security Program begins
to recognize and reward farmers’ investments in these vital assets, both
prospectively and retroactively. Despite official U.S. nonparticipation in
global carbon-trading regimes, the farm-sector benefits of reduced CO2
emissions are efficiently available from private trading regimes now
emerging around the world.
An average of
450,000 agricultural
jobs (farm proprietors
and employees)
have been lost each
decade since 1970.
But once cellulosic
biofuel technology
lowers the cost of
ethanol, the game
changes fundamentally. It’s no longer
about who can lobby
best for crop subsidies, but about strategic investment in a
new and more secure
rural-based domestic
fuels infrastructure
that competes,
without subsidy,
even with today’s
subsidized gasoline.
684. Indeed, the recent
Energy Bill foundered
partly on a bipartisan
gridlock between the rural
states and the urban areas
on the degree of ethanol
subsidies that the urban
areas would be forced to
pay (Coon 2004; Taylor &
Van Doren 2003).
685. According to AUS Consultants, Inc., increasing ethanol output to 5b gal/y by 2012 would reduce crude oil imports by 1.6b bbl cumulatively,
cut the U.S. trade deficit by $34 billion, and eliminate $10.6 billion of direct government payments to farmers. Environmentally, net CO2 emissions are cut by
78% for biodiesel (Schumacher, Van Gerpen, & Adams 2004) and 68% for ethanol (assuming a E85 fuel blend—GM et al. 2001, Fig. 3.6).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
163
Implementation
Crafting an effective energy strategy: Restoring farming, ranching, and forest economies
The renewable fuels
farming system can
provide farmers with
total revenues of
~$500–900/acre-y
not counting any of
the $100–500/acre-y
carbon offset revenue
from windpower
production. This
income level compares favorably with
gross revenues generated from such traditional crops as corn
($325/acre-y), wheat
($118), and soybeans
($281). Together these
pure-net income
streams could
quadruple pretax
net farm income.
The carbon offsets from biofuel production could potentially be worth
$26–128/acre-y, assuming carbon prices of $10–50/tonne carbon.686
Lease payments for windpower production (worth ~$1,000–1,600/acre-y
to the turbine owner) can provide an additional $50–80/acre-y for areas
with Class 4 winds or better, which are widespread on the Great Plains.687
The carbon offsets from the wind turbines’ output would be worth an
additional $100–500/acre-y,688 some of which could probably be captured
by the landowner in a competitive market. Finally, soil carbon storage
can add a further $3–15/acre-y.689
Collectively, the renewable fuels farming system can provide farmers
with total revenues of ~$500–900/acre-y (Table 4), not counting any of the
$100–500/acre-y carbon offset revenue from windpower production (in
case the windpower developer captures all of that benefit). This income
level compares favorably with gross revenues generated from such traditional crops as corn ($325/acre-y), wheat ($118), and soybeans ($281),690
all of which require far costlier inputs and operations than harvesting
perennial switchgrass. Better yet, about $80–220/acre-y of carbon revenues
and wind royalties (again ignoring carbon offsets from windpower) incur
no out-of-pocket costs to the farmer or rancher—they’re third-party payments for byproducts of the main farming activity—and together these
pure-net income streams could quadruple pretax net farm income, which
averaged only $43/acre-y in 2002 (p. 108). Obviously not every farmer can
capture all these kinds of income, and exact values will differ widely, but
equally obviously this approach holds promise of major improvements in
farm economics, and analogously for ranch and forest operations.
Table 4: Potential farm revenue and net income from clean energy and carbon offsets (assuming carbon trading at ~$10–50/tonne carbon).
Windpower net income counts only royalties from wind-turbine siting, on the assumption that the carbon offset value from the wind
electricity’s displacing fossil fuels would be captured by the windfarm developer, not the landowner.
Carbon offset value,
2000 $/acre-year
Value to farmer,
2000 $/acre-year
(switchgrass)
26–128
457–765 including
>26–128 net
windpower
1,000–1,600
100–500
50–80 net
soil carbon
–
3–15
3–15 net
1,431–2,237
129–643
510–860
Product
biofuels
total
Product value,
2000 $/acre-year
431–637
686. Assuming a crop yield of 7 ton/acre-y and an ethanol conversion rate of 180 gal ethanol/ton, 22 barrels of crude equivalent per acre could be replaced
each year. Using a 0.85 kg/L density for crude and an 84% carbon content results in 2.56 tonne/acre-y carbon savings.
687. The American Wind Energy Association estimates that income to farmers from wind rights amount to $50–80/acre. Annual income from a single 1.5-MW
wind turbine will be perhaps $3–4,000 per year (depending upon how much electricity is generated), and only ~0.5 acre will be used to site the turbine within
the ~50–75 acres dictated by spacing requirements (AWEA 2004a).
688. Assuming a wind energy output of 40 MWh/acre-y, displacing coal-fired generation that releases 0.25 tonne carbon/MWh, yields a savings of 10 tonne
carbon/acre-y or $100–500/acre-y (at $10–50/tonne carbon).
689. Assuming sequestration of 0.3 tonne carbon/acre-y (Swisher 1997; Swisher et al. 1997) worth $10–50/TC yields $3–15/acre-y of pretax net revenue to the farmer.
690. Revenue per acre was calculated by multiplying partly irrigated crop yield in bushels/acre (USDA/NASS 2002, Table 33—Specified Crops Harvested Yield
per Acre Irrigated and Non-Irrigated) by the estimated 2003/04 price per bushel (USDA 2004).
164
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Crafting an effective energy strategy: Restoring farming, ranching, and forest economies
The new demand for bioenergy crops, plus the other new agricultural
revenues, should provide a major boost to rural employment. Estimates
vary, but could be as high as 780,000 new agricultural sector jobs from
biofuels production based on national production of 24.9 billion gal/y of
ethanol.691 Our 2025 State of the Art ethanol volume, 57.7 bgal/y, is 2.3
times that big, corresponding at our 23% lower job intensity to 1.45 million added jobs—clearly a rough approximation, but equivalent to nearly
half of total U.S. agricultural employment today.
Rural employment
could be as high as
780,000 new agricultural sector jobs from
biofuels production.
Table 4 summarizes how five potential new revenue streams to the agricultural economy—biofuels, windpower, and carbon offsets from both of
these energy sources, as well as from soil carbon storage—could exceed
those from conventional farming, while reducing input costs and local
impacts on land and water resources. Thus reducing oil dependence, fossil-fuel consumption, and greenhouse gas emissions can help American
agriculture to become more profitable as well as more sustainable.
There are broader whole-system benefits as well. From a government
perspective, a cost-effective biofuels program would allow the government
to restructure the billions of dollars in farm subsidies that are paid each
year—$11.7 billion in 2002 (2000 $) including $1.9 billion to corn, $1.7 billion to the Conservation Reserve Program, and $0.7 billion to soybeans.692
Potentially, some 5% of the annual expenditure, or $540 million per year,
could be eliminated or substantially reduced.693 (Of course, if greater net
farm income were simply offset by reduced subsidies, farmers would
be no better off in aggregate, but there could be significant distributional
effects, since current subsidies are widely considered to favor big over
small operations.) Finally, of course, having a greater mix of energy
supplies from domestic sources increases national energy security: the
~4 Mbbl/d of cost-effective biofuels (p. 103) would be the all-American
equivalent of about half a Saudi Arabia, but undisruptable and inexhaustible. As the Apollo Project’s advocates put it, would we rather
depend on the Mideast or the Midwest?
The ~4 Mbbl/d of cost-effective biofuels would be
the all-American equivalent of about half a Saudi Arabia,
but undisruptable and inexhaustible—
would we rather depend on the Mideast or the Midwest?
691. Urbanchuk 2003. Assuming 57 bgal/y of ethanol production from our SOA scenario at its feedstock cost of $11.3 b/y,
and the agricultural “Job Creation Multiple” of 28 jobs created per million dollars of revenue (Laitner, Goodman, & Krier
1994) implies the creation of 743,000 jobs—lower than the Urbanchuk estimate, but significant. Alternatively and probably
more accurately, an average 40 Mgal/y ethanol plant adds 694 jobs throughout the economy (RFA 2004);
our proposed output would need 1,300 such plants, creating 957,000 jobs.
692. Environmental Working Group 2002.
693. Urbanchuk 2001. That study assumes that loan deficiency payments will be eliminated and government payments will
be reduced, therefore lowering direct payments to farmers by $7.8 billion during 2002–16, or $540M/y.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
165
Implementation
Sooner or later,
the market will
provide superefficient cars,
trucks, and airplanes.
The opening moves
have already been
made to determine
who will sell them.
The game will
move swiftly.
Foreign rivals now
enjoy a lead from
their national
policies.
The United States
must catch up—
not by copycat efficiency mandates,
but by policy innovations tailored to
the U.S. business
conditions we’ve just
described.
Neither
competitive forces
nor oil-market
concerns will wait
for business-asusual.
If we don’t act soon,
the invisible hand
will become the invisible fist
The market will ultimately deliver more efficient cars, trucks, and airplanes. Indeed, it already has begun to do so with the introduction of
doubled-efficiency hybrid vehicles in the past few years and Boeing’s 7E7
in 2007–08. But will it be U.S. or foreign manufacturers who reap the benefit of this market transition? If U.S. business and political leaders are not
decisive now, we could end up replacing imported oil with imported cars
and airplanes. And while competitive global markets will ultimately
bring efficient fleets to U.S. buyers, this could take so long that the U.S.
automotive sector exits just as its successors enter. Let’s therefore recap
the business and policy challenge before we move on to the policies that
can help business leaders to address it timely.
The fundamental business problem in U.S. automaking is the disparate
financial strength of the Big Three and their competitors. The Big Three
simply do not have the ability to invest quickly in an entirely new and
properly broad product line, even at the lower product development and
production costs we’ve described, without giving up one of the strategic
R&D programs they’re already conducting. Of course the Big Three can,
should, and do incorporate lighter and stronger parts into their existing
product lines and new platforms, but that’s incremental change, while the
coming global shift in automotive technology and markets requires radical and rapid change. Neither competitive forces nor the oil-market concerns discussed on pp. 8–25 will wait for business-as-usual.
By contrast, Toyota, and to a lesser degree Honda, Nissan, and others,
have the financial strength and the business strategy to adopt holistic
bundles of new technology rapidly in new product lines. They continue
to have the benefits of their domestic markets, access to lower-cost capital, and the keiretsu supply-chain system, plus the lower-cost, high-quality
lean manufacturing system made famous by Toyota and the support of
their domestic and regional markets.
Korean automakers should not be overlooked. They’ve gained more points
of global market share in the 1990s than any other country and are considered a disruptive force in the automobile market.694 They have succeeded in
the way Toyota pioneered, by entry into the lowest-profit-margin segment
of the business, the small car market. Like Toyota, their strategy is to leverage that position into the higher-margin segments. What better way than
with breakthrough technology? The same opportunity is doubtless occurring to India’s burgeoning automakers.
694. Christensen & Raynor 2003, p.61.
166
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
If we don't act soon, the invisible hand will become the invisible fist
The Chinese have the potential to build a new automobile industry from
scratch, and they show every sign of preparing to leapfrog the West.
To appreciate the power of a massive domestic market with patriotic buying behavior, overseen by a farsighted central policy, consider China’s
growth in the technology sector. Over the past five years, China has sustained a 40% annual growth rate in high-tech exports, and now manufactures 35% of the world’s cellphones. China has passed the U.S. to become
the world’s largest mobile phone market (as noted on p. 6, note 46, it has
more cellphone users than the U.S. has people), and China Mobile and
China Unicom are the numbers one and three largest mobile carriers
globally.695
The explosive growth of Chinese automotive manufacturing capability
is already positioning it as a tremendous threat to U.S. automakers
(p. 135).696 In 2003 alone, sales of Chinese-made vehicles in China grew by
more than one-third (p. 2). Not only are Chinese companies investing in
automaking, but General Motors plans to invest $3 billion in its Chinese
operations over the next three years,697 and GM’s competitors are similarly
ambitious. Yet China’s intention to serve more than its home and regional
market could hardly be plainer: the State Development and Reform
Commission’s 2 June 2004 white paper (p. 135) says, “Before 2010, our
country will become an important vehicle making nation, locally made
products will basically satisfy domestic demand, and we will enter the
international market in a big way.”698 Shanghai Automotive Industry
Group already hopes to partner with MG Rover Group Ltd., buy
Ssangyong (Korea’s fourth-biggest automaker), and become one of the
world’s top ten automakers.698a
China, Europe, and Japan all have the advantage of technology-forcing
regulations that stimulate demand for new, more efficient cars. These rules
appear to reflect some rivalry, and the gap between these regulations and
U.S. rules is widening. China’s 2004 white paper requires that the “average
fuel consumption for newly assembled passenger vehicles by the year
2010 will be reduced by at least 15 percent compared to the level of 2003”;
this applies better-than-U.S. minimum standards to every new vehicle (not
just their average) and bars used-car imports.699 European automakers, as
an alternative to legislated standards, have voluntarily committed to the
equivalent of 39 mpg by 2008 (25% lower fuel intensity) and are considering 46 mpg (another 15% intensity drop) by 2012.700 New legislation in
Japan requires 23% fuel-economy gains during 1995–2010, to levels up to
44 mpg depending on vehicle class, via the “Top Runner” program (p. 45,
note 225) committing all new vehicles to approach the most fuel-frugal
vehicle in their class.701 Canada’s Climate Action Plan, with 2004 bipartisan
endorsement, promises 25% mpg gains by 2010 despite the close integration of its auto industry with that of the U.S.
Implementation
China, Europe,
and Japan all have
the advantage of
technology-forcing
regulations that
stimulate demand
for new,
more efficient cars.
695. Cai 2003; Fu 2004.
696. China Economic Net
2004.
697. Zhengzheng 2004;
Agencies 2004a.
698. Blanchard 2004.
698a. Wonacott 2004.
699. Bradsher 2003.
700. European automakers
are implementing a voluntary standard of 140 grams
of CO2 per km by 2008 and
negotiating a standard
likely to be 120 gCO2/km for
2012. European emissions
averaged 166 gCO2/km in
2002, down 11% from 186 in
1995. That progress reflects
mainly dieselization,
but Chancellor Schröder,
at the request of German
automakers, recently
asked the EU to accelerate
dramatically the Euro-5
standards that will reduce
fine-particle emissions.
701. Fulton 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
167
Implementation
China’s intention
to serve more than its
home and regional
market could hardly
be plainer:
“Before 2010, our
country will become
an important vehicle
making nation,
locally made products will basically
satisfy domestic
demand,
and we will enter the
international market
in a big way.”
If we don't act soon, the invisible hand will become the invisible fist
In striking contrast, no increase is contemplated in the U.S. 27.5-mpg
average car standard—passed in 1975 and first effective in MY1984—
and the 2003 1.5-mpg increase in U.S. light-truck standards to just 22.2
mpg for MY2007 will yield little or no actual gain.702 This growing gap
hardly bodes well for U.S. exports to an increasingly integrated global
market. As we’ll show next, efficiency standards aren’t the only or even
the best way to improve fleet efficiency, but if, for whatever cause, new
American cars’ efficiency continues to stagnate, then most foreign buyers,
and an increasing share of U.S. buyers, simply won’t want them.
However, a key competitive opportunity remains as unexploited by foreign as by U.S. policymakers. While the U.S., Japan, and Europe are all
providing generous R&D funding for the fuel cell as the next-generation
engine, all of their official programs give far more limited attention to the
near-term practical potential for a lighter, stronger platform and body to
get to fuel cells sooner and solve the hydrogen storage problem (p. 233),
while saving far more gasoline and diesel fuel meanwhile. The United
States has at least as great technological breadth and depth in advanced
materials as Japan and Europe, and a more entrepreneurial flexible market system for rapidly exploiting that capability.
Capital quickly migrates to those companies that demonstrate the ability
to earn superior returns. The invisible hand of the market, as described by
Adam Smith, was an early attempt to describe the movement of
capital in society. The speed of capital migration has accelerated of late,
so the invisible hand will feel more like the invisible fist when it hits
those automakers that fall behind in the competitive race for the nextgeneration vehicle.
The entire extra
business investment
needed to reach
~77% market capture
in 2025 by superefficient cars and
light trucks and
~70% by efficient
heavy trucks—
some $70 billion
spread over a decade
or two—is what the
United States now
spends directly
every seven months
to buy foreign oil
that’s largely wasted.
Do U.S. business leaders and policymakers really want to wait for the
market to deliver the next generation of efficient mobility via leisurely
meandering, or should we enact the few critical strategy and policy shifts
that would transform the market more quickly and potentially save and
revitalize the U.S. transportation manufacturing sector? As we debate this,
consider that the entire extra business investment needed to reach ~77%
market capture in 2025 by superefficient cars and light trucks and ~70%
by efficient heavy trucks—some $70 billion spread over a decade or two—
is what the United States now spends directly every seven months to buy
foreign oil that’s largely wasted.
702. That’s because the higher standard would be roughly offset by a proposed extension of the current law’s “flex-fuel”
loophole, which credits ethanol-blend-capable vehicles with fuel savings (up to 1.2 mpg for MY1993–2004) even if they
never use alternative fuel blends (other than normal oxygenates); very few actually do. Sales in the CAFE-exempt >8,500-lb
GWVR category, such as Hummer, have also increased since Oak Ridge National Laboratory found 5.8 million such exempt
“heavy light trucks” on the road at the end of 1999, accounting for 9% of light trucks’ total fuel use (NHTSA, undated).
168
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Crafting coherent supportive policies
Government’s role in implementation
Since the journey beyond oil is not costly but profitable, free enterprise
will lead it in pursuit of greater profit, stronger competitive advantage,
and lower business risk. But as the previous 41 pages have shown,
the existing policy framework is indifferent, inconsistent,703 rife with
distortions, and scarcely focused on accelerating fundamental innovation.
Current policy encourages only routine rates of spontaneous market
progress that are too little and too late for national needs, maintaining
a business environment that is far too risky for key American industries
in a world full of resourceful and determined rivals.
We said on p. 1 that governments should steer, not row. They should set
the rules of a fair and free marketplace, not choose its outcomes nor
supplant the private firms that do the work. But the preceding business
cases compel us to conclude that some judicious, coherent, and supportive
steering in the direction of accelerating technological leapfrogs can turn
serious business risks and societal challenges into major national benefits.
To make the transition faster, cheaper, and surer, policy must accelerate
oil efficiency by:
Since phasing out oil
is not costly but
profitable, business
will lead it for profit.
Coherent public
policy can steer in
the right direction by
expanding choice
and adoption, and
speeding development and production,
of advanced-technology vehicles.
It can also foster
alternative energy
supplies, reward
what we want rather
than the opposite,
and enhance
competition between
different modes of
transport and ways to
need them less.
• shifting customer choice strongly toward advanced-technology vehicles
while expanding freedom of choice, so enterprise can deliver uncompromised, indeed enhanced, service and value in ways that also produce public goods;
• reducing the risk of retooling, retraining, new capacity, and widely
beneficial R&D;
• supporting private investment in innovative domestic energy supply
infrastructure;
• purging perverse incentives (i.e., stop doing dumb things); and
• expanding alternatives for fair access, at honest prices, to all competing
modes of access and mobility, including smart development that
requires less travel.
As we describe innovative ways to achieve these goals, please bear in
mind a bedrock need for firms and governments at all levels: systematic
and comprehensive “barrier-busting” is vital to let people and companies
respond efficiently to the price signals they already see. That alchemy of
obstacles into opportunities—or as Interface, Inc.’s Chairman Ray C.
Anderson says, “stumbling-blocks into stepping-stones”—is described
more fully elsewhere,704 in a broader context than just oil and transportation. It underlies all the specifics that follow.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
703. Most strikingly in the
huge tax deductions
offered for “business” use
of the heaviest and least
efficient SUVs—a heavily
marketed tax break that
can cover most of the cost
of, say, a dentist’s buying
a Hummer.
704. Lovins & Lovins 1997,
especially pp. 11–20.
169
Implementation
Crafting coherent supportive policies: Government’s role in implementation
Cynicism about governments is
millennia old and often deserved
(Box 17). But to the widespread
assumption that not much about
oil use can or will change, and
perhaps even less if government
gets involved, we suggest five
replies:
Did anyone notice
that national water
productivity in 2000
was only 2.36 times
its 1975 value,
and per-capita withdrawals were down
26% (while per-capita
energy use rose 5%)?
Did we have a
terrible problem of
needing to muster
national will to suffer
the pain of achieving
this miracle?
170
• Throughout history, in such
crises as World War II, democracies have risen magnificently to
meet and defeat great dangers.
Such mobilizations, including
U.S. automakers’ six-month
crash-conversion to military
production (p. 1), spring from
extraordinary leadership. But
creeping crises short of this cataclysmic character can also be
mastered by less conspicuous
and more routine ways of just
paying attention.
• For example, with astonishing
benefit and virtually no disruption, the United States has
already demonstrated extremely
rapid oil and energy savings
during 1977–85, led by government but implemented largely
by the private sector (pp. 7–8).
The savings called for here are
considerably slower than those
that were already achieved
with so little fuss that, less than
two decades later, they’re
already nearly forgotten.
17: Gridlock as Usual
according to
Thucycides, ca.
431–404 BCE
In his 24 June 1964 speech in
Frankfurt’s Paulskirche cathedral,
President Kennedy paraphrased
and quoted a noted Athenian
soldier and historian: 705
Thucydides reported that the
Peloponnesians and their allies were
mighty in battle but handicapped by
their policy-making body—in which,
he related, “each [faction or state]
presses its own end...which generally
results in no action at all.…They
devote more time to the prosecution
of their own purposes than to consideration of the general welfare—each
supposes that no harm will come of
his own neglect, that it is the business
of another to do this and that—and so,
as each separately entertains the
same illusion, the common cause
imperceptibly decays.”
705. Kennedy 1963. Cebrowski (1999) used a
similar quotation when President of the Naval War
College. The quoted portion is apparently adapted
from Thucydides 431 BCE (the 1910 Crawley translation). The original (including the paraphrased part)
in The Peloponnesian War 1.141.6–7 is characteristically terse:
[6] µάχ˙ µ¢ν γὰρ µιᾷ πρÚς ἅπαντας
῞Ελληνας δυνατο‹ ΠελοποννÆσιοι κα‹ ο
ξʵµαχοι ἀντισχε›ν, πολεµε›ν δ¢ µØ πρÚς
ıµοαν ἀντιπαρασκευØν ἀδÊνατοι, ˜ταν
µÆτε βουλευτηρƒ •ν‹ χρ≈µενοι
παραχρ∞µά τι Ùξ°ως §πιτελ«σι πάντες
τε $σÒψηφοι ˆντες κα‹ οÈχ ıµÒφυλοι τÚ
§φ’ •αυτÚν ßκαστος σπεÊδ˙: §ξ …ν φιλε›
µηδ¢ν §πιτελ¢ς γγνεσθαι. [7] κα‹ γὰρ ο
µ¢ν …ς µάλιστα τιµωρÆσασθα τινα
βοÊλονται, ο δ¢ …ς ¥κιστα τὰ ο$κε›α
φθε›ραι. χρÒνιο τε ξυνιÒντες+ §ν βραχε› µ¢ν
µορƒ σκοποËσ τι τ«ν κοιν«ν, τ“ δ¢
πλ°ονι τὰ ο$κε›α πράσσουσι, κα‹ ßκαστος
οÈ παρὰ τØν •αυτοË ἀµ°λειαν ο‡εται βλάψειν, µ°λειν δ¢ τινι κα‹ ἄλλƒ Íπ¢ρ •αυτοË
τι προÛδε›ν, Àστε τ“ αÈτ“ ÍπÚ
ἁπάντων $δᾳ δοξάσµατι λανθάνειν τÚ
κοινÚν ἁθρÒον φθειρÒµενον.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Crafting coherent supportive policies: Government’s role in implementation
• Energy isn’t the only example of such rapid change (Fig. 35). During
1975–2000, U.S. withdrawals of water per dollar of GDP fell at an
average compounded rate of 3.4%/y—faster than the ~2.7%/y which
we propose for 2000–25 oil efficiency. To be sure, though energy and
water use both depend on composition of GDP, the two resources are
quite different, and they’re saved by different actors and means.706 But
among their similarities is that usage of both appeared to rise in lockstep with GDP until the mid-1970s, when that correlation was deliberately broken, causing steady declines in usage not only per dollar of
GDP but also per capita. Did anyone notice that national water productivity in 2000 was only 2.36 times its 1975 value, and per-capita
withdrawals were down 26% (while per-capita energy use rose 5%)?
Did we have a terrible problem of needing to muster national will to
suffer the pain of achieving this miracle? Does anyone doubt it was a
good idea? Can you imagine our water situation today if we hadn’t
done it? This water-efficiency revolution came mainly from millions of
sensible corporate and personal choices within a slowly improving
policy framework, such as starting to price water on the margin,
reduce subsidies to its use, educate users, and make markets in saved
water. These policy shifts drove many business shifts, such as paying
attention to water, improving technical designs, and bringing waterthrifty products to the marketplace (encouraged by national standards
for plumbing fixtures).707 The result is astounding, and this invisible
chapter in U.S. resource-productivity history is far from over.
Figure 35: Water productivity, 1950–2000. During 1950–2000, U.S. GDP more than quintupled and population nearly doubled, yet total water withdrawals 708 were lower in 2000 than in 1980. Per dollar of GDP,
water withdrawals were 59% lower in 2000 than in 1950, and 62% lower than their peak in 1955. To most
Americans, this stunning progress in water productivity was completely invisible, quietly driven by private
decisions supported by mainly decentralized public policy. Yet without it, we’d now be withdrawing 2.5
times as much water as we actually do, and most of the country would be in a severe and continual water
crisis. Making today’s oil problems fade away over the next few decades can have similar dynamics,
advantages, and ultimately invisibility if we gently steer the system in the right direction.
6
real GDP
4
population
water withdrawals
3
water/capita
1.00
1
1.00
1.23
1.26
1.34
1.50
1.62
1.60
1.35
1.26
1.20
0.57
0.49
0.41
1995
2
1990
index (1950 = 1.0)
5
1.38
1.07
1.07
0.96
0.97
0.96
0.84
0.65
Source: RMI analysis from USGS data (Hutson et al. 2004).
year
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
2000
1985
1980
1975
1970
1965
1960
1955
1950
0
water/GDP
Five replies
to the assumption that
oil use can’t or won’t
change that much
(continued)
706. The biggest water savings were in agricultural
irrigation and thermal
power plants (due mainly to
fewer new nuclear plants,
which need more cooling
water per kWh, and more
combined-cycle plants).
However, important gains
were also made in industry,
buildings, and landscaping
irrigation.
707. Rocky Mountain
Institute helped to speed
this process 16 years ago
by simply publishing a
thorough catalog of littleknown market offerings.
The industry soon took over
that informational function.
708. Hutson et al. 2004.
171
Implementation
Crafting coherent supportive policies: Government’s role in implementation
Five replies
to the assumption that
oil use can’t or won’t
change that much
(continued):
• The same success of unobtrusive but pervasive policy shifts is evident
in increased electric efficiency at the state level. For example, aligning
utility with customer interests and adopting building and appliance
efficiency standards early held per-capita use of electricity virtually
flat since the mid-1970s in California, while it rose by more than half
in the rest of the country. The savings helped California achieve faster
economic growth.709 Similar results were achieved in New England
households after 1989. Vermont households held electricity use steady
during 1974–90, then lowered it by an eighth. Nobody felt hampered;
they simply wrung better services at lower cost from less electricity
and more brains.
709. Contrary to a widespread myth, California’s
2000–01 power shortages
were not due to unexpectedly soaring demand, nor to
a shortage of generating
capacity, which continued
to expand through the
1990s but in more costeffective nonutility and
decentralized forms (Lovins
2001a). Mostly, the shortages were due to a perfect
storm of poorly designed
restructuring and rational
(if sometimes illegal) market response to its unintended perverse incentives.
710. Or at least very close
to it if other considerations
make a theoretically secondbest solution preferable.
711. Democracy, as law
professor Harold Green
remarked, is based not on
the expectation of truth but
on the certainty of error.
Cyberneticists express the
same idea when they say
that systems without feedback are stupid by definition. Policies gang oft agley;
as FDR remarked in the
Depression, we can’t know
that this or that will work,
but we must try. With the
humble awareness that
unintended consequences
accompany most policy
interventions, we nonetheless think the ideas suggested next merit consideration,
testing, and refinement in
the forge of experience.
172
• As we’ll show, the shifts we’re proposing in automaking are slower
than those the industry actually achieved in the 1970s and 1980s, less
capital-intensive, and probably a good deal more profitable. And they
require less invention than such previous requirements as the Clean
Air Act, which was passed—with justified faith in American ingenuity—before the industry had even developed the catalytic converter.
For implementation methods, policymakers can draw on decades of experience—chiefly in Europe, Japan, and North America, but increasingly
also in developing countries whose abundant needs and talents often
incubate clusters of innovation. However, the U.S. policy toolkit, despite
some past successes, has been mired in twenty years of trench warfare,
and is ripe for renewal. In general, U.S. car companies favor fuel taxes
while U.S. oil companies favor car-efficiency standards, and these two
titanic lobbies have fought each other to a draw since 1975. It’s time to
seek fresh solutions.
After briefly reviewing the lessons of those two dominant post-1973
policy tools, we’ll propose modern alternatives that meet clear criteria.
To make our own biases explicit, we believe that:
• Government should protect public health, safety, and security;
enhance equity, choice, and competition; align incentives so efficient
markets produce, and don’t destroy, public goods; and remove unnecessary obstacles to more perfect markets.
• Energy policy instruments should be company-, entrant-, and technology-neutral, innovation-accelerating, economically efficient,710
evidence-shaped, market- and results-based, fair, nondiscriminatory,
easily understood, transparent, and attractive across the political
spectrum.
• The instruments to be preferred are revenue-neutral (paying for themselves with no net flow to or from the Treasury), are technology-forcing, reward continuous improvement, learn quickly from rich and
effective feedback,711 and self-destruct after their job is done.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Government’s role in implementation
Since all five of this study’s coauthors have studied or taught economics,
yet we’re all field practitioners—not armchair theorists who lie awake at
night wondering whether what works in practice can possibly work in
theory—we feel compelled to add another comment, hinted at on pp.
25–28. The market is a powerful and efficient tool for the short-term allocation of scarce resources. But markets also have limits.712 They’re not good
at long-term, and especially at intergenerational, allocation; generally
aren’t concerned with distributional equity (markets are meant to be efficient, not fair); may not tell us how much is enough; reveal cost and perhaps value but not values; and can’t substitute for politics, ethics, or faith.
Markets are wonderful tools, and we apply and refine them extensively,
but they can’t do everything, and their purpose is far from the whole
purpose of a human being.
With this understanding, let’s first explore why and how the goals of
traditional tax- and mandate-based U.S. policy instruments can be better
achieved without them.
Fuel taxes
The United States taxes gasoline and diesel fuel at some of the lowest
rates in the world, manyfold lower than Europe and Japan do. Much
higher taxes are the most obvious, economically doctrinaire,713 and politically difficult way to signal the true social costs of buying and burning oil
and the public goods of using less of it. Such conservatives as University
of Chicago Nobel economist Professor Gary Becker, 714 Hudson Institute
director of economic policy studies Dr. Irwin Stelzer,715 and columnist
Charles Krauthammer, 716 as well as commentators ranging from environmental writer Gregg Easterbrook to Ford Motor Company Chairman Bill
Ford, Jr., all favor this solution.717 (Taxing aviation fuel at all—it’s tax-free
nearly worldwide, thanks to an intricate network of hundreds of treaties
that were meant to promote air travel and succeeded with a vengeance—
would shrink this giant distortion between transportation modes.718
But let’s focus here on road vehicles’ fuel.)
Implementation
The U.S. policy toolkit,
despite some past
successes, has been
mired in twenty years
of trench warfare,
and is ripe for renewal.
America underprices
mobility fuels,
but their economically
correct pricing,
though useful, isn’t
effective, sufficient,
or necessary for
encouraging the
purchase of efficient
vehicles. It’s more
useful for a different
purpose—signaling
the social cost of
traveling.
712. The austere preconditions for a perfect market—e.g., perfect information about the future, perfectly accurate price signals, perfect competition,
no monopoly or monopsony, no unemployment or
underemployment of any
resource, no transaction
cost, no subsidy, no
taxes—clearly don’t
describe the world any of
us inhabit. The differences
between that theoretical
world and the real world
are what makes business
innovation possible and
profitable.
713. Since Arthur Cecil Pigou proposed them in 1918. He is generally credited with the now-common distinction between private and social marginal products and costs, and with the idea that the resulting market failures can be corrected by “internalizing the externalities” through taxes and subsidies. (Ronald
Coase in 1960 showed that absent transaction costs, private transactions on broadened property rights could do the same thing without governments.)
714. Becker 2004. Dr. Becker estimates that a 50¢/gal “terrorist protection tax” on gasoline would cut gasoline use by ~10%, and suggests it be used to
increase stockpiles.
715. Stelzer 2004. Dr. Stelzer particularly likes taxes on oil imports (for which, as we noted at p. 13, note 97, there is legal authority). Of course, the U.S.
already taxes its oil imports—via DoD’s part of our tax bills (p. 20)—and so does OPEC.
716. Krauthammer 2004.
717. Gross 2004.
718. The Chicago Treaty regime’s prohibition of aviation fuel taxes doesn’t bar CO2 emissions fees. A new EU-U.S. “open skies” aviation agreement wasn’t
signed in mid-2004 partly because many EU countries propose such fees, but the U.S. wants them charged only on EU airlines, not on U.S. airlines operating
within Europe (a view echoed by the UK). A 188-nation Montréal conference will try again in September 2004 to soften U.S. intransigence.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
173
Implementation
Crafting coherent supportive policies: Government’s role in implementation: Fuel taxes
We consider the
high gasoline taxes
favored by most
major U.S. allies and
trading partners to be
theoretically sound
and generally helpful,
but the weakest
possible signal to buy
an efficient vehicle,
insufficient,
and nonessential.
With a big caveat, we consider the high gasoline taxes favored by most
major U.S. allies and trading partners to be theoretically sound and generally helpful in improving economic efficiency, because fuel prices omit
important costs to society, and prices that tell the truth are always better
than prices that lie about what things really cost. High gasoline taxes
needn’t be regressive if immediately recycled into corresponding cuts in
other taxes, especially the most regressive ones such as payroll taxes.
Fuel taxes could take various forms, including carbon charges from capand-trade CO 2 markets,719 though these wouldn’t send a big price signal:
$25 per tonne of carbon is equivalent to only 6¢/gal, well within normal
short-term price fluctuations. So what’s the caveat? High gasoline taxes
also have serious defects beyond their obvious political challenges.720
They are:
719. Former Federal
Reserve Chairman Paul
Volcker and former Council
of Economic Advisors
chairman Martin Feldstein
have proposed that carbon
taxes (equivalent to efficient trading) be used to
cut federal budget deficits,
and Ford Motor Company
has reportedly supported
using a carbon tax to
replace part of California’s
sales tax (Morris 1994,
note 1). Europe is shifting
toward taxing less the
things we want more of
(like income and jobs)
while taxing more the
things we want less of
(like depletion and pollution). See Repetto et al.
1992; PETRAS 2002; and the
green tax shifting papers
at www.redefiningprogress.
org/publications/
720. An enduringly bizarre
aspect of American political life is that Americans
seem to prefer OPEC ministers rather than members
of the U.S. Congress to be
the ones raising their
prices at the pump, and
OPEC countries (and unintentionally, in small part,
terrorists) rather than the
U.S. Treasury to receive the
revenues.
• the weakest possible signal to buy an efficient vehicle.721 Since U.S.
taxed gasoline is only about an eighth of the total cost of owning and
running a car, this price signal is first diluted ~7:1 and then shrunk to
insignificance by consumer discount rates—which are so high that the
difference between a 30- and a 40-mpg car, though consequential for
society, will seem to the short-sighted buyer to be worth only about
the price of a set of floor mats, hardly worth the hassle of laborious
learning and negotiations;722
• insufficient (if it were sufficient, the many foreign countries with
$4–5/gallon gasoline prices would be driving State of the Art vehicles
by now, but they aren’t 723—in fact transportation is the fastest-growing
CO2 source in Europe);
• nonessential and hence not worth arguing about, because there are
even better ways to encourage people to choose efficient vehicles.
The main virtue of higher gasoline taxes would be in reducing miles driven after the car is bought.724 This effect, though weak, is clearly observable;
but we’ll propose alternative ways to achieve that goal too, without
increasing net cashflow to the Treasury. So without denying the sound
economic principle of proper pricing, we think there are more creative,
politically palatable, and effective ways to signal the value of efficient
vehicles—especially at the point of purchase where that decision is
focused. We’ll return to the most important such option—feebates—as
soon as we’ve argued that efficiency standards, too, are no longer the best
policy choice for America.
721. IEA 2001. The graph on p. 22, for example, shows that around 1997, new cars were about an eighth more efficient in Britain and Germany than in the U.S.
(where they were larger, p. 27), but were less efficient in Japan (where they were generally smaller) despite its severalfold higher fuel taxes.
722. Greene 1997, pp. 4–7. This near-indifference to efficiency by customers, in turn, causes producers to see in such incentives a low incentive but a high
investment risk (Greene 1997, p. 8).
723–724. See next page.
174
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Government’s role in implementation
Standards, mandates, and quotas
Before we similarly note the limitations of efficiency standards, let’s also
note their historical context. In the decade after Congress enacted Corporate Average Fuel Economy (CAFE) standards in 1975 as one of the least
controversial elements of the Energy Conservation and Policy Act,725 U.S.
oil use dropped 7% and oil imports dropped 23% while GDP grew 37%.
Informed by detailed analyses and hearings, the standards were carefully
set at levels that could be met cost-effectively (or very nearly so) with
straightforwardly available technologies. They therefore did not—as many
feared and some still claim—prove costly, inefficient, or unsafe.726 In hindsight, CAFE standards and their milder light-truck equivalent were largely
responsible for nearly doubling car efficiency and increasing light-truck
efficiency by more than half in their first decade. (A gas-guzzler tax on
cars—though not on light trucks—and a few years of high oil prices after
the 1979 spike helped too.) CAFE was the cornerstone of America’s OPECbusting oil savings in 1977–85 (pp. 7–8), and helped to avert even graver
erosion of U.S. automakers’ market share by more efficient Japanese
imports (p. 140).
The 1975 law required CAFE standards to enforce “maximum feasible
fuel economy,” considering technological feasibility, economic practicality,
fuel-economy effect of other standards, and the nation’s need to conserve
energy. (Lawmakers have lately tried to add other criteria.) The standards
were thus meant to evolve as automotive technology became more energy-efficient—as indeed it did by one-third during 1981–2003 (p. 8), at
prices the public paid. Yet during 1975–2003, the standards weren’t raised
at all, contrary to the strong wishes of three-fourths of Americans in
1995.727 Instead, the standards have become such anathema to most U.S.
automakers that they’ve been broadly changed only twice728—weakened
5.5% for cars in MY1986–88 (after world oil prices fell by half, due substantially to CAFE-induced efficiency gains), then strengthened 7% for
light trucks during MY2005–2007. Thus the 21.0-mpg light-truck standard
that Congress voted in 1975 to take effect in MY1985 got delayed 20
years.729 During 1995–2000, Congress expressly forbade even considering
any CAFE increases. Now the responsible agency, the National Highway
Implementation
Foreign countries and
even states are pulling
dangerously far ahead
of U.S. efficiency
standards, but there
are even more economically efficient,
effective, flexible,
and speedy ways to
shift to fuel-frugal,
high-performance,
market-leading
vehicles.
Roughly 85–90% of
the world’s energy
twenty years from
now will be used by
products not yet manufactured, and many
of these will be made
in China.
723. In general, they’re simply driving less, partly because they often have more sensible land-use and better public-transportation alternatives. The EU is
having to introduce new non-price policy instruments and voluntary agreements (backed up by a tacit threat of mandates) to bring next-generation vehicle
efficiency to market, because gasoline taxes near the limits of political tolerance aren’t enough to achieve this. As the IEA (2001, p. 57) puts it, “Even big
changes to fuel prices may not have much additional effect on vehicle choices.” To mention a simple empirical observation, the state of Hawai‘i has gasoline
prices 50¢/gal higher than the mainland U.S., yet its fleet mix is not appreciably more efficient than the U.S. national average.
724. In the short run, partly due to rigidities in land-use patterns and alternative transportation options, vehicle travel is remarkably unresponsive to cost.
The International Energy Agency concludes that “a 10% increase in fuel prices results in only a 1–3% decrease in travel” (IEA 2001, p. 12).
725. Bamberger 1999. The key votes approving CAFE’s authorizing statute (EPCA) were 300–103 in the House and 58–40 in the Senate (U.S. Congress 1975).
726. Greene 1997.
727. Greene 1997, Table 3.
728. Plus three single-year weakenings—by 7% in 1979 for the 2WD light-truck standard in MY1981, by 7% in 1984 for the average light truck in MY1985, and
by 4% in 1988 for the car standard in MY1989 (Bamberger 2003).
729. Bamberger 2003, Table 1.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
175
Implementation
Crafting coherent supportive policies: Government’s role in implementation: Standards, mandates, and quotas
Traffic Safety Administration (NHTSA)—while making some commendable
safety improvements and perhaps reducing some of the “gaming” that has
seriously undercut CAFE’s goals—is proposing a weight-based rewrite that
may weaken this foundation of America’s oil-saving success while undermining future auto exports (pp. 58, 206–208). In response to this federal
inaction and erosion, many states are starting to act on their own.730
730. The leader, California,
is making rules in 2004 to
implement “maximum
feasible cost-effective
reduction” in light vehicles’
CO2 emissions starting with
MY2009; the 2002 authorizing law excludes mandatory trip reduction, any additional fees or taxes, or banning any vehicle category.
The growing move for
coastal Eastern states,
challenged by ozone nonattainment, to adopt California’s standards could give
its CO2 rule strong market
influence if it survives
threatened legal challenges. Of course, as noted
earlier (note 685), a CO2
standard can be met by
switching to low- or nofossil-carbon fuels, or
partly by such other means
as CFC refrigerant abatement, rather than solely
by better mpg.
731. LBNL, undated.
Nonetheless, many still
oppose and block costeffective standards on
apparently ideological
grounds. For example, a
recent rollback of federal
air-conditioner standards
threatened to waste 1.34
TCF of gas over the next
25 years, but was reversed
in court. DOE is still delaying standards for major
categories of gas-fired furnaces, boilers, and water
heaters that could save
5.5 TCF/y over the next
20 years.
Undeterred, many other nations also continue to tighten national technical
standards for fuel efficiency (p. 167), as WTO rules permit so long as the
standard is rationally based and nondiscriminatory. Standards have a long
record of effectiveness not only in vehicle efficiency and emissions (such as
the Clean Air Act) but also in raising the energy efficiency of buildings and
appliances. U.S. appliance standards, for example, have already saved 40
billion watts (GW) of peak electricity demand in refrigerator/freezers and
135 GW in air conditioners, compared with, say, the 61 GW of load lost in
the 14 August 2003 Northeast blackout. So far, the federal government has
spent ~$2 per household devising and implementing nine residential
appliance efficiency standards to combat two market failures—customers’
taking a far shorter view of future energy savings than society does, and
split incentives (most appliances are bought by landlords, homebuilders,
and housing authorities who won’t pay the utility bills). By 2020, that
$2-per-household investment in minimum standards will have stimulated
each household to spend an additional $950 on efficiency, thereby saving
$2,400—a $150 billion net saving to the national economy.731 The rise of
efficiency standards for everything from lights to motors and buildings to
cars in rapidly growing economies like China’s is fortunate, since roughly
85–90% of the world’s energy twenty years from now will be used by
products not yet manufactured, and many of these will be made in China.
Standards also have their limits. Imposing mandates can be less efficient
than market mechanisms.732 Standards aren’t as fine-grained or transparent
to customers as methods that translate vehicles’ efficiency directly into
their sticker price. Standards are readily gamed, as CAFE has been to
near-perfection. Moreover, absent political consensus to take seriously
the CAFE law’s provision for updating, the standards, once set, are static.
And once standards are met, they can halt backsliding but foster no further innovation, because they include no reward for beating the standard.
Standards also don’t take account of different firms’ differing markets,
circumstances, and ability to comply at different costs; in contrast, tradable
permits for sulfur, nitrogen-oxide, and lead pollution have achieved very
large pollution reductions and dollar savings simultaneously, surpassing
their promoters’ fondest hopes.733
732. A typical exposition of this thesis is Kleit 2002. However, Greene (1997) argues that a combination of standards and fuel taxes works better than either
alone, due to consumers’ “satisficing” behavior, producers’ risk aversion, and the sluggishness and partial oligopoly of vehicle-efficiency markets.
733. Stavins 2001. The most famous market-based program, EPA’s SOx allowance trading system, has cut SOx by 6.5 million tons since 1980 at an estimated
cost saving of $1b/y. One of the first U.S. tradable permit programs, for lead in the 1980s, met its environmental goals, saved ~$250M/y, and provided measurable incentives for technology development and diffusion. See also EPA 2002.
176
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Government’s role in implementation: Standards, mandates, and quotas
Automakers also complain that CAFE standards can push customers
toward smaller cars that meet their needs less well, might be less safe,
and are far less profitable to make. This hardly seems a serious problem,
since during 1975–2003 the market share of small cars fell by 15% while
that of SUVs rose 22%, nearly surpassing them in total sales.734 Furthermore, under CAFE automakers can sell and customers can buy whatever
vehicles they wish, subject to a small ($55/car-mpg) penalty paid, at least
in theory, by noncompliant manufacturers.735 Some, including the UAW,
assert that CAFE has tilted competition to favor imported cars, although
this is hard to square with historical evidence.736 Nonetheless, the critique
persists that CAFE distorts customer choice.
Regardless of the merits, which are complex and controversial, the parties
to the CAFE dispute, chiefly auto and environmental lobbies, are so firmly dug into polarized positions, and the subject has become so politically
toxic, that any end-run around it is likely to be preferable. Our policy
approach not only achieves the oil-saving and competitive-strategy goals
better; it may not even require federal legislation (which is only one way to
implement policy), and far from raising tax burdens, it should markedly
reduce the federal budget deficit.
These attributes are desirable not only to evade gridlock, but also because
the legislative sausage-factory is prone to opaque back-room negotiations
that too often turn temporary, narrow, pump-priming interventions—
especially tax expenditures—into such eternal entitlements that some U.S.
energy sectors have been milking subsidies for more than a half-century.737
(Depletion allowances were introduced in 1918 to spur energy output for
World War I.738) Moreover, while carefully targeted tax expenditures can
be effective,739 their design often rewards spending, not results, and may
discriminate against anyone who doesn’t pay taxes.
For all these reasons, we prefer the innovative, market-oriented policies
described next. They are flexible enough to offer choices between federal
and regional or state implementation, so that the federal government can
lead, follow, or get out of the way. They offer similarly wide choice for
administrative vs. legislative implementation. And they offer, we believe, a
trans-ideological appeal conspicuously absent from the proposals that have
deadlocked federal energy policy for the past few years. We’ll start with
the five major steps that the federal government (other than DoD) should
take—and that state governments could take absent federal leadership—
to reduce oil use in light vehicles. We’ll next list an assortment of other federal opportunities, and then move to those distinctively available to states,
the military, and civil society. Finally, we’ll assess budgetary impacts, which
turn out to be gratifyingly positive. We won’t formally analyze “free riders”
(people who’d have done as hoped even without incentives) because we
don’t think this is an important problem, especially when offset by “free
drivers” (early adopters who follow incentivized adopters’ good example).
Implementation
The parties to the
CAFE dispute, chiefly
auto and environmental lobbies, are so
firmly dug into
polarized positions,
and the subject has
become so politically
toxic, that any endrun around it is likely
to be preferable.
Our policy approach
not only achieves the
oil-saving and competitive-strategy
goals better; it may
not even require
federal legislation
(which is only one
way to implement
policy), and far from
raising tax burdens,
it should markedly
reduce the federal
budget deficit .
734. EPA 2003, p. 32.
735. These civil penalties
since 1983 total >$0.5 billion.
European automakers
regularly pay ~$1–20+M/y as
a cost of doing business in
the U.S. Asian makers and
DaimlerChrysler have never
incurred the penalty. Ford
and GM have on occasion,
but have never had to pay
it thanks to retroactive
loopholes.
736. Greene 1997, pp. 30–32;
NAS/NRC 2001, p. 6-2. The
Academy recommended
eliminating both the dual-fleet
rule and the dual-fuel allowance now in CAFE rules.
737. EIA 1999a.
738. Lovins & Heede 1981, p.28.
739. Datta & Grasso 1998.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
177
Implementation
Crafting coherent supportive policies
Better policies
to guide market
evolution toward both
business success
and national security
goals can cut government expenditure
and avoid new
federal legislation.
Synergistic reforms
can change federal
policy from a brake
to an accelerator,
and profitably lift
advanced vehicles’
market share
from zero to more
than three-fourths.
Federal policy recommendations for light vehicles
740. EFC 2003, Summary
of Recommendations:
Transportation Working
Group, pp. 1–3 and
App. A: Working Group
Reports, pp. 1–14.
178
Our policy and technology recommendations for light vehicles closely
match those published in 2003 by a task force of the Energy Future Coalition. The Transportation Working Group was chaired by Dennis R. Minano
(GM’s Vice President for Environment and Energy 1993–2003) and comprised the Big Three automakers, a Tier One supplier, the United Auto
Workers, Shell, and three leading technology- and business-oriented environmental groups.740 After “an intensive review of the best ways” to bring
advanced automotive technologies “more quickly and in greater volumes
into the marketplace,” the group agreed on four central recommendations:
Initiative 1: Establish incentives for manufacturing and purchasing advanced
technology vehicles.
Already, U.S. manufacturers are preparing to produce and market a range of
more efficient advanced technology vehicles. But without external incentives,
the transition to the broad manufacture and consumer acceptance of vehicles
with advanced fuel-saving technologies will be slow, too slow to help significantly on the issues of [oil] dependence and climate in the necessary timeframe.
To accelerate the deployment of these vehicles into the marketplace, the Working
Group recommends a mix of manufacturing and consumer incentives that will
partially offset the higher purchase price of these vehicles and reduce manufacturers’ capital needs as they retool to produce these vehicles. These incentives
should, though, be sharply targeted. The Group recommends that, to qualify for
either incentive, a vehicle must, at a minimum, meet performance criteria relating to fuel use.
Initiative 2: Ensure the availability of clean fuels for advanced vehicles
[i.e., cleaner gasoline and diesel fuel, biofuels such as ethanol and biodiesel,
and hydrogen].
Initiative 3: Invest in the aggressive development of fuel cells.
Initiative 4: Reduce vehicle miles traveled.
Since Initiatives 2–3 are treated elsewhere in this report, and Initiative 4 is
beyond its scope (p. 38), we focus here on the Work Group’s Initiative 1.
It aims at “getting millions, not thousands, of advanced technology vehicles
on the road quickly” via “significant [consumer and manufacturer] incentives that primarily focus on lowering consumer costs for advanced fuelsaving technology vehicles, as well as incentives for U.S. manufacture of
these vehicles” via facility conversion credits and complementary support
via government fleet purchasing. The purpose and effect of our proposals
exactly mirrors those of this consensus, but we suggest some improvements in detailed structure to improve efficiency and avoid tax burdens.
For example, we favor mechanisms that reward producers for making and
selling efficient vehicles (and their components) rather than for spending
money; similarly for customers, we suggest revenue-neutral feebates scaled
to efficiency rather than tax credits scaled to expenditures. Despite such differences in detail, we’re encouraged by the striking parallels between the
task force’s consensus based on its diverse stakeholder interests and our
independent recommendations based on the preceding business-model
analysis and public-policy goals.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
“[W]ithout external incentives, the transition to the broad manufacture
Let’s start with our conclusions
and consumer acceptance of vehicles with advanced fuel-saving
about how a suite of policies can
technologies will be slow, too slow to help significantly on the issues of
greatly accelerate the introduction
[oil] dependence and climate in the necessary timeframe.
and adoption of advanced vehicle
To accelerate the deployment of these vehicles into the marketplace,
technologies, as summarized in the
the Working Group recommends a mix of manufacturing and consumer
following graphical double-spread
incentives that will partially offset the higher purchase price of these
(Fig. 36). Our strategy uses eight
vehicles and reduce manufacturers’ capital needs
policies to shift demand toward
as they retool to produce these vehicles.”
efficient vehicles and to mitigate
— Energy Future Coalition Transportation Working Group,
manufacturers’ risk-aversion and
comprising
illiquidity when they retool to meet
DaimlerChrysler,
that demand. These policies were
Ford, GM,
simulated and refined via a model of light-vehicle sales and stocks that
Visteon,
we constructed for this report, based on the conceptual structure and price
UAW, Shell,
elasticities from a 1995 U.S. Department of Energy study.741 Our treatment
Natural Resources
of feebates is consistent with an important peer-reviewed 2004 paper
Defense Council,
by noted researchers at DOE and its Oak Ridge National Laboratory; 742
Union of
although their model is more complex and is structured differently,
Concerned Scientists,
and
it responds similarly to feebates. Box 18 explains the model structure
World Resources
and key assumptions that yield the results in Fig. 36. Starting on p. 186,
Institute
we’ll explore each of the policies simulated in Fig. 36.
As a reminder: Conventional Wisdom vehicles are like today’s steel, gasoline-engine vehicles, but with consistent use of the straightforward and
moderate improvements—nearly all in the propulsion system—that some
market vehicles already include (such as variable valve timing).
Conventional Wisdom vehicles embody the kinds of incremental improvements analyzed in the National Research Council’s 2001 study, summarized above on pp. 49–51. As we said on p. 69, CW vehicles save 27% of
the fuel used by the light vehicles assumed in EIA’s forecast of U.S. oil use
to 2025. In contrast, State of the Art vehicles are ultralight hybrids, which
cost more and save much more. The two are compared in cost and savings
in Fig. 21, p. 66. The efficiency of today’s relatively heavy hybrids falls in
between these two categories and is not specifically modeled in our study,
so those vehicles can be considered either a proof of the conservatism of
CW vehicles or a stepping-stone to SOA vehicles, which add ultralight
construction and other refinements to today’s best hybrid powertrains.
741. Davis et al. 1995.
742. Greene et al. 2004.
743–745. See box 17 on p. 182.
CAUTION: ENTERING CALCULATIONAL THICKET.
The next two pages simply and graphically summarize our main light-vehicle policies and their simulated effects.
Figure 36 shows each policy’s incremental effects on the sales of new vehicles and the on-the-road vehicle stock.
Box 18 explains more technically how we simulated these effects.
If you don’t need to know how our model works, we suggest you read just pages 181–182, then skip to p. 185.
And if you don’t need an in-depth understanding of our proposed federal policies, you can skip all the way to p. 204
or even to p. 207.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
179
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
Figure 36: How RMI’s eight proposed policies, described next, can stimulate demand for and accelerate production of State of the Art
light vehicles (green)—and, as a side-effect, of Conventional Wisdom (yellow) ones too—compared with EIA’s forecast (made in January
2004 and extending to 2025) of relatively inefficient vehicles (red). For each successive policy step, market share of new vehicles is
graphed on the left, and the light-vehicle fleet on the road (the vehicle stock) on the right. On the left, the Conventional Wisdom and State
of the Art market shares in 2025 are shown in yellow and green, respectively. On the right, the boldface roman-type number is the lightvehicle petroleum product saving compared with EIA’s forecast for 2025; the boldface italic number is the 2004 net present value (NPV)
for vehicles purchased between 2005–25 (over the 14-y vehicle life) of customers’ net savings, i.e. the fuel saving at EIA’s forecast retail
gasoline price minus the vehicles’ extra pretax retail price; and the lightface italic numbers are the 2025/2050 average mpg of the fleet.
40
180
2035
2030
2025
2020
2015
2010
0
2005
20
2050
2045
2040
2035
2030
2020
2015
2025
2035
2030
2025
2050
60
d. This scenario also includes
the low-income scrap-andreplace program. It’s not a
big oil-saver, but it’s vital for
equitable social development,
and creates a profitable new
million-vehicle-a-year market.
2050
% of sales
80
2045
100
2010
400
350
300
250
200
150
100
50
0
99%, 0%
2045
5
0
5
0
5
0
0
5
20
2040
40
2040
60
2.7, $295 , 26.4/28.3
2.8, $323 , 26.7/28.4
2035
% of sales
80
400
350
300
250
200
150
100
50
0
2030
100
Let’s Get Started scenario
c. The first and most important
Coherent Engagement policy
is feebates, starting here at the
basic $1,000/0.01-gpm (gallons
per mile) level that effectively
lets buyers see 14, not 3, years’
fuel savings. This nearly drives
EIAmobiles out of the market.
2025
94%, 0%
2020
2035
2030
2025
2020
2015
2010
0
2005
20
2020
40
2015
60
2015
% of sales
80
1.6, $179 , 23.7/24.8
2010
100
400
350
300
250
200
150
100
50
0
2010
53%, 0%
Drift scenario
b. Allowing incremental 27%
efficiency gains beyond EIA’s
technology assumptions leads
those Conventional Wisdom
vehicles to capture half the
market (declining slightly later
as EIAmobiles improve)
because CW (assumed to be
static) pays back in a few years.
To distinguish the effects of
State of the Art vehicles,
we don’t show them until case e.
2005
2035
2030
2025
2020
2015
2010
0
2005
20
2005
40
2005
60
million vehicles
% of sales
80
400
350
300
250
200
150
100
50
0
million vehicles
100
STOCK (million vehicles)
2025 oil saving 0 Mbbl/d
NPV customer saving $0 billion
fleet mpg 2025/2050: 21.2 /—
million vehicles
2025 share: 0%, 0%
a. EIA forecast without
Conventional Wisdom
or State of the Art technology
EIA projects that all new
light vehicles are “EIAmobiles.”
They include 5% (1.1 million)
new hybrids in 2025, but those
get only 34 mpg for cars and
27 for light trucks (worse than
today’s hybrids). EIA’s average
new light vehicle in 2025
is thus only 0.5 mpg more
efficient than it was in 1987.
million vehicles
NEW SALES (CW, SOA )
percentage of EIA forecast sales;
last two graphs on p. 181
exceed 100% due to scrappage
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Figure 36 (continued)
NEW SALES (CW, SOA )
STOCK (million vehicles)
2050
2045
2045
2035
2030
2025
400
350
300
250
200
150
100
50
0
2020
400
350
300
250
200
150
100
50
0
2040
2050
2035
2030
2025
2020
2015
2010
0
2005
20
2040
40
2035
60
2030
% of sales
80
3.4, $319, 28.6/52.1
2025
100
Mobilization scenario
f. Combining the full (albeit
static) technology portfolio—
State of the Art vehicles
plus the Conventional
Wisdom ones pulled along
in their wake—with Coherent
Engagement policies transforms the market, starting
with basic feebates, the most
important policy.
2020
51%, 44%
2015
2035
2030
2025
2020
2015
2010
0
2005
20
2015
40
2010
60
Drift scenario revisited
e. If our no-policy (Drift)
scenario also assumes State
of the Art vehicles (although
they’d come to market slowly
[at best] absent demanddriving policies), they’d still
gain a modest early-adopter
market share.
2005
% of sales
80
2010
million vehicles
400
350
300
250
200
150
100
50
0
million vehicles
100
2005
million vehicles
million vehicles
400
350
300
250
200
150
100
50
0
million vehicles
1.9, $190, 24.4/29.9
400
350
300
250
200
150
100
50
0
34%, 19%
40
2035
2030
2025
20
i. Despite these demand
stimuli, retooling lags constrain State of the Art
vehicles’ entry into the slowturnover U.S. fleet. Adding
smart government procurement and Golden and
Platinum Carrot competitions
cuts three years off the time
for State of the Art vehicles
to reach 10% market share.
0
5
4.5, $391, 32.5/61.0
2050
60
2020
5
80
2015
2045
110
100
2010
0
27%, 77%
2005
2040
0
2035
20
2030
40
2025
60
2020
% of sales
80
2015
h. Increasing the feebate
slope to $2,000/0.01 gpm,
to start to count public goods,
increases SOA market
capture, mainly in the later
years after plants have had
time to retool.
2010
110
100
% of sales
0
3.6, $360, 29.3/58.9
54%, 49%
0
5
5
0
5
0
5
0
0
5
20
0
40
5
60
2005
% of sales
80
g. Adding the low-income
scrap-and-replace program
mainly sells more Conventional Wisdom vehicles
because they’re available
earlier. As State of the Art
vehicles become available too,
feebates speed their adoption.
0
100
5
3.5, $319, 28.9/52.4
55%, 44%
Source: RMI analysis. Methodology in Box 17, pp. 182–185; policy descriptions on pp. 186–206.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
181
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
18: Modeling the effects of policy
on light vehicle sales and stocks
To explore the effects of our suggested policies,
RMI constructed a peer-reviewed dynamic
model of U.S. light vehicle fleet sales and
stocks. It’s fully described and downloadable
(Technical Annex, Ch. 21), structurally simple,
flexible, easy to use and alter, yet able to capture essential behaviors with reasonable accuracy. No model can exactly predict complex
systems, but we believe this model—unique, to
our knowledge—provides a sound and transparent basis for assessing our policy proposals.
The feebate affects customers’ benefit/cost
ratio, calculated as the rebate for an efficient
car (Box 18) plus the present value of its 3-year
fuel saving—discounted at 5%/y (p. 39)—all
divided by the vehicle’s incremental pretax retail
price. Benefit/cost ratios shift purchasing
behavior according to DOE’s historic price elasticities, which are assumed to be identical for
SOA and CW vehicles. From purchasing behavior, RMI’s model calculates vehicle stocks, fuel
savings, and their retail-customer net value.
Our model tracks car and light truck fleets by
annual age cohort. It matches EIA’s 2001–25
fleet forecast (EIA 2004) within ~3%. RMI’s
model estimates how public policy affects both
demand and supply. Eight policies were summarized graphically on the previous two pages,
and will be described in the following text. The
five policies that stimulate sales are feebates,
low-income scrap-and-replace, federal fleet
procurement, its coordination with a “Golden
Carrot,” and a “Platinum Carrot” competition to
pull further innovation into the market. The three
policies that stimulate production of efficient
vehicles are manufacturer conversion incentives (modeled as federal loan guarantees,
though they could take other forms), military
R&D investment, and early military procurement
(whose production investments in turn create
capacity for production for the public market).
Feebates are by far the most important, and
could achieve all the benefits of the demandside policies but ~5–10 years later.
We assume EIA’s projections of sales mix and
vehicle-miles traveled in each year to 2025. Thus
our assumptions about how many and what
kinds of vehicles people buy, how much they’re
driven, and the sales volumes, prices, and efficiencies of conventional vehicles are all EIA’s.
Our CW and SOA vehicles’ cost and mpg come
from Boxes 7–10 (pp. 62–73). Our model endogenously computes only how quickly those two
vehicle categories get adopted, assuming introduction in 2005 and 2010 respectively (Fig. 37a,
p. 183). Our CW and SOA vehicles achieve EIA’s
projected 2025 0–60-mph acceleration time in
every year, so before 2025, they’re peppier than
average (by a gradually decreasing amount), but
we don’t assume they’ll therefore sell better.
Some 90–95%743 of the efficiency gain from feebates comes from manufacturers’ making the
same vehicles more efficient, but our model conservatively counts no fuel savings from the other
5–10%, which comes from customers’ buying a
different mix of vehicles. We model technological progress under all policies as the gradual
adoption of State of the Art vehicles, all bought
instead of Conventional Wisdom vehicles rather
Each year, our model uses the previous year’s
sales of cars and of light trucks to adjust the
“pivot point” (Box 18) to keep feebates revenueneutral, and then to adjust production so manufacturers make more of what has just sold well.
182
(continued on next page)
743. Davis et al. 1995; Greene et al. 2004 (for the ~95% figure).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Box 18: Modeling the effects of policy on light vehicle sales and stocks (continued)
744. Survival rates for RMI’s cohort model are based on Table 3.9 and
3.10 of ORNL 2003. VMT is based in Table 3.6 and 3.7, scaled up to
EIA’s projections through 2025, then extrapolated linearly.
745. RMI’s model is annually dynamic, whereas Greene et al. (2004)
model a static pivot point and assess a one-year snapshot ~10–15
years in the future. Nonetheless, when we tested our model by introducing NRC vehicles, it predicted mpgs 5% above to 10% below those
predicted by Greene et al. (Technical Annex, Ch. 21). As for oil savings, Patterson, Steiner, & Singh (2002) show that a weighted-average
vehicle stock average of ~53 mpg corresponds to refined-product use
of 7.3 Mbbl/d, which scales to ~5.4 Mbbl/d using RMI’s vehicles. RMI’s
model finds 5.1 Mbbl/d—reasonable agreement given the differences
in assumed technology costs and policies. Of course, these models
have different purposes: RMI’s calculates from consumer preferences, while Patterson, Steiner, & Singh’s only asks what mpg would
be needed to achieve a given oil saving.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
2050
2045
2040
2035
2030
2025
2020
2015
100
90
80
70
60
50
40
30
20
10
0
2010
Conventional Wisdom
State of the Art —with technology procurement
State of the Art —with financing support
2005
Adoption in each year equals demand times
retooling fraction (Fig. 37a); CW and SOA vehicles can’t be adopted faster than manufacturers
retool to produce them. Thus if everyone wanted State of the Art vehicles but only 50% of
automakers’ capacity had been retooled to produce them, then only half the demand would be
met; the rest would default to Conventional
Wisdom vehicles (or, to the extent they too
weren’t yet available or demanded, to EIA’s inefficient base-case vehicles). Conversely, if manufacturers had retooled to make every vehicle
State of the Art , their sales would nonetheless
be limited by demand. Thus our model matches
annual demand with supply insofar as this supply is available within the practical constraints
of retooling.
Figure 37a: Assumed evolution of how quickly
Conventional Wisdom and State of the Art vehicles could
potentially capture the indicated share of the U.S. market
for new light vehicles if every efficient car that could be
produced were demanded. The yellow CW curve starts
earlier and rises faster because it simply spreads
fleetwide the incremental, familiar technologies already
used today in some market platforms. The green SOA
curves assume the retooling support (probably in the form
of loan guarantees) described on pp. 203–204; the dashed
green curve shows further acceleration by the technology
procurement and “Golden Carrot” policies (pp. 199–200).
percent retooled
than EIAmobiles (thus conservatively minimizing
oil savings). We assume that scrapped vehicles
have the same mpg as other vehicles contemporaneously retired. We assume that all light vehicles last 14 years,744 although carbon-composite
SOA vehicles should last far longer (and could
be designed for fuel-cell retrofits). We assume
that a given policy yields only one unique outcome, different policies yield different outcomes,
and an economically indifferent manufacturer
will split choices 50/50. Our model’s output
agrees well with two DOE model results.745
year
Source: RMI analysis.
Our best judgment of the practical constraints
on retooling and retraining is shown in Fig. 37a,
based on standard logistic s-curves and reflecting a moderate mix of policies that include federal action to relieve automakers’ and suppliers’
financial constraints as described later on pp.
203–204. By “priming the pump,” our other proposed policies should affect not just demand
(their main aim) but also retooling rates, and
this is shown schematically in Fig. 37a’s dashed
green line as a three-year acceleration of the
initial “takeoff” phase of making State of the Art
vehicles.
(continued on next page)
183
Implementation
Box 18: Modeling the effects of policy on light vehicle sales and stocks (continued)
How soon could manufacturers start producing
State of the Art vehicles? For advanced-composite versions, Fig. 37b illustrates how this
could plausibly occur in 2010 based on a concerted effort launched in 2005 by one or more
major market players. Automakers don’t traditionally start designing platforms whose production methods aren’t long perfected and practiced, but they’ve lately been doing such parallel
development with certain light-metal production
innovations where prudent fallback positions
were available. In this case, ultralight steel (pp.
55, 67) offers such a “backstop” using conventional fabrication methods. But the dynamics
shown in Fig. 37b appear realistic in light of six
main factors: current development status (pp.
56–57), U.S. technological depth, market imperatives, the speed and power of modern virtual
design techniques, the relative simplicity of tooling to make composite autobodies, and the
technological shortcuts already embedded in
our assumed designs (e.g., using cosmetic exte-
rior panels so composite structures don’t need
Class A finish). To make Fig. 37b feasible basically requires at least one decisive automaker
or Tier One supplier, and there are many candidates. The 2010 date might even be surpassed
by the most aggressive competitors.
Different policies match different technological
worldviews, so we considered presenting
different policy packages matched to the
Conventional Wisdom and State of the Art
vehicle technologies. We chose instead to
describe a single policy package focused on
accelerating State of the Art vehicles to
achieve radical and profitable oil savings. In
that world, Conventional Wisdom vehicles’
incremental oil savings are also available in the
marketplace, but are pulled along as a byproduct of accelerating State of the Art vehicle
adoption. To be sure, the production of
Conventional Wisdom vehicles starts sooner in
(continued on next page)
Figure 37b: Assumed schematic schedule for launching 50,000/y production of State of the Art vehicles in 2010 (0.25% of that year’s
light-vehicle sales), based on advanced-composite production technology demonstrated in mid-2004 at 1×1-m scale (or on BMW
technology, pp. 56–57), then perfected for large-scale use through a larger industry/federal R&D effort (pp. 204–206). Delays would
either correspondingly delay SOA vehicles’ introduction and fuel savings or require a switch to light-steel substitutes (pp. 55, 67),
but we consider this schedule feasible.
refine process for manufacturing carbon-fiber composite structure
platform development
build & equip greenfield factory
fabricate tooling
pre-production prototypes, homologation, & limited production
2005
2006
2007
2008
year
2009
2010
Source: RMI analysis in consultation with industry experts.
184
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
Implementation
A major lesson of Fig. 36 is that despite a strong portfolio of both technologies and policies, the stock of light vehicles on the road in 2025 gets
only 32.5 mpg, compared with 21.2 mpg in EIA’s projection. By 2050,
the 32.5 has risen to 61.0 mpg, reflecting the maturation of stock turnover.
These very long lead times are inherent in the dynamics of the enormous,
long-lived, slow-moving vehicle stock, the lead times of retooling, and the
supertanker-like momentum of the whole automotive system. The policy
lesson is clear: start quickly and work aggressively to get very efficient
vehicles on the road as soon as possible in large numbers. It’s like the
king who, in the legends of several cultures, told his gardener he wanted
to plant a tree. “Sire, it will take a hundred years to grow,” replied the
gardener. “Then,” said the king, “we must plant it this afternoon!”
So what are the policy instruments that would accelerate adoption and
production of State of the Art vehicles as dramatically as Fig. 36 shows?
What are the key insights from that policy modeling? As mentioned in
Box 18, the first and most important instrument is feebates, because they
stimulate consumer demand for, and industry supply of, ever more efficient vehicles. Feebates aim to ensure that the most efficient vehicles at
any given time are the highest-margin vehicles, so manufacturers are
eager to develop and sell them. Policy must also help to accelerate supply; this requires the conversion of existing factories to build State of the
Art vehicles to meet that burgeoning demand, and may also require new
factories. An integrated suite of policies is needed here: early military science and technology investment to coalesce, strengthen, and expand the
advanced materials sector; guaranteed civilian and military government
procurement of advanced vehicles in the early years (a “Golden Carrot”)
Box 18: Modeling the effects of policy on light vehicle sales and stocks
(continued)
this scenario, and since they are much cheaper,
they’re adopted more quickly (Fig. 37b). Although
they save only 39% as much fuel as State of the
Art , Conventional Wisdom vehicles therefore
yield most of the total oil savings from all light
vehicles by 2025. However, State of the Art vehicles, after the delay caused by their later start
and slower retooling, complete their takeover
not long after 2025, ultimately saving far more oil.
And in any event, the red (inefficient) vehicles
that underpin EIA’s forecast are completely
squeezed out of the market by 2020 under our full
Coherent Engagement policy scenario (Fig. 36i).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Even the impressive oil saving shown in Fig. 36i
is conservative, mainly because we assume
no technological progress beyond 2004 (p. 37)
and no adoption of fuel cells. In practice, since
State of the Art vehicles greatly accelerate
the market entry of affordable fuel cells (pp.
233–234), those should achieve production scale
at competitive cost late in the period. They
would save about two-fifths more energy than
shown for the appropriate fraction of State of
the Art gasoline-hybrid vehicles and would
entirely displace hybrids’ remaining oil use,
since the hydrogen for the fuel cells (pp.
227–242) would come from saved natural gas or
renewables or both.
185
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
Rebates for buying
efficient vehicles,
paid for by fees on
inefficient ones,
can expand customer
choice, create both
consumer and producer surpluses,
cut fuel costs by hundreds of billions of
dollars in net present
value, and save
millions of barrels
of oil every day.
Incentives would favor
efficient vehicles,
not smaller ones,
and the automakers
who make the best
vehicles, whatever
their market segment.
so as to build industry manufacturing experience and volume more rapidly; a “Platinum Carrot” prize competition to reward early mass-market
adoption of even more advanced vehicles; and throughout, especially
at first, the financial support to prime the pump by providing the U.S.
automaking industry with the financial liquidity needed to restructure its
manufacturing capacity. In addition, we suggest a low-income lease-andscrap program designed to help reduce the burdens on the welfare system and further expand the market for efficient new vehicles. The rest of
this section details how these policies would work, as well as some additional options that should be thoughtfully considered.
Feebates
The centerpiece of our policy recommendation is the “feebate,” which
provides a rebate for or levies a fee on each new vehicle depending on its
efficiency.746 Buyers of new light vehicles that exceed a certain annually
defined fuel economy benchmark, called the “pivot point,” would receive
a rebate to be subtracted from the purchase price. The amount of the
rebate would depend on how much the vehicle’s fuel economy exceeds
the pivot point for vehicles of that size. Conversely, buyers of new vehicles with fuel economies lower
than the pivot point for vehicles of
19: How feebates work
that size would pay a corresponding surcharge on their purchase
Feebates lower the prices of efficient vehicles, so people
price. The feebates we propose
buy more of them, and raise the prices of inefficient
are revenue-neutral, with no net
vehicles, so people buy fewer of them. Each year, the fees
flow of dollars into or out of the
pay for the rebates (plus the minor administrative costs747).
Treasury. Instead, the fees paid by
Consider, for example, a feebate of $1,000 per 0.01 gpm,
buyers of less efficient vehicles
with a pivot point of 23 miles per gallon (0.043 gpm) for the
(which impose social costs) would
midsized SUV class. A Nissan Pathfinder getting 18 miles
be used to pay the rebates to buyers of more efficient vehicles
per gallon (1/18 mpg = 0.056 gpm), is 0.13 gpm worse
(which save social costs), with a
than this benchmark value or “pivot point,” so Pathfinder
tiny bit left over to pay the feeincurs a $1,300 fee. Ford’s new Escape hybrid SUV gets
bates’ administrative costs.
(let’s suppose) 36 miles per gallon, or 0.028 gpm—0.015
Feebates are typically described as
gpm better than the pivot point—so it would earn a $1,500
a dollar value for every gallon per
rebate. These changes in the vehicles’ retail prices are
mile difference from the mpg pivot
typically smaller than the sales incentives that most
point (see Box 19)—not mile per
automakers now offer out of their own profit margins,
gallon, since the goal is to save
and are economically about equivalent to the hybrid tax
gallons in a linearly proportional
credits offered by some states such as Colorado. In this
manner (i.e., all gallons saved are
example, the feebate would be revenue-neutral if slightly
equally valuable).748 As the fleet
more Pathfinders than Escape hybrids were sold.
747. We haven’t explicitly accounted for administrative costs because in well-run programs, based on extensive utility experience, they should be very small—well within
analytic uncertainties. But they must be budgeted and paid for.
746. Greene et al. 2004; Koomey & Rosenfeld 1990.
See also note 636.
748. See next page.
186
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Feebates
becomes more efficient, the pivot point would gradually shift 749 toward
lower fuel intensity (higher mpg), eventually surpassing Conventional
Wisdom vehicles. If we ever ran out of worthwhile technologies, we could
declare victory and stop.
U.S. feebates should be structured to be revenue-neutral, technologyneutral, and neutral as to vehicle size so as to enhance and not distort
customer choice. We therefore suggest having the feebate system apply the
same slope ($/gpm) to each and every new light vehicle without exception, but with a separate pivot point for each size class (measured by
interior volume or areal footprint as the best metric of customer utility—
not by weight).750 Size-class-based feebates will preserve the competitive
position of each automaker regardless of where in the market it concentrates its offerings, and thus put no U.S. automaker at any disadvantage.751
Feebates apply to each vehicle, not to the average of all sales by each manufacturer. However, treating feebates by size class also nearly eliminates
the potential for shifting customer choice between classes, and avoids
complaints about interference with freedom of choice. Whatever size of
vehicle you prefer, you have a choice whether to get one that’s more or
less efficient, with the corresponding rebate or fee attached. So a business
buying a powerful, high-capacity Class 2b truck would pay no more
(and may pay less) for a fuel-efficient version of that vehicle with feebates
than for an inefficient version in a system like today’s, without feebates—
and that business will also pay far less for fuel. All gain, no pain.
Unlike standards, feebates reward and propel continuous improvement.752
Feebates are provided to the customer, but they actually incentivize the
manufacturer to incorporate energy efficiency improvements that cost less
than the feebate “reward,” thus maintaining an attractive retail pricepoint.
Unlike fuel taxation, feebates directly signal the value of efficiency to the
buyer at the time and place of choosing the vehicle. Feebates are a continuous mathematical function, and are completely transparent and predictable
to manufacturers and customers, making feebates more efficient than
standards. They’re also less prone to “gaming,” although any system will
be gamed somehow. And feebates make money for everyone: under a
$1,000/0.01-gpm feebate, automakers’ net sales would increase by nearly
Implementation
The centerpiece
of our policy recommendation is the
“feebate,” which
provides a rebate for
or levies a fee on
each new vehicle
depending on its
efficiency, with no
net flow of dollars
into or out of
the Treasury.
Feebates should be
structured to be
revenue-neutral,
technology-neutral,
and neutral as to
vehicle size so as to
enhance
and not distort
customer choice.
Unlike standards,
feebates reward and
propel continuous
improvement.
748. Basing linear feebate
slopes (rates) on mpg
instead of gpm would introduce serious distortions,
since a 1-mpg increase in a
15-mpg vehicle saves 10
times more gallons per mile
than a 1-mpg increase in a
48-mpg vehicle. However, it
is conventional and appropriate to express the pivot
point in mpg for easy comparison with federal efficiency labels.
749. If each year’s setting of the pivot point were purely retrospective, the continuous improvement of market offerings would prevent revenue-neutrality.
The calculation should therefore be based on estimated sales mix for the year about to occur, much as EPA now does when estimating each model year’s
sales-weighted efficiency, or as state utility regulators do when calculating electricity and gas tariffs for a future “test year.” Although sales projections
even a year ahead are always uncertain, the industry’s production lead times would allow each model’s mpg to be closely estimated far enough in advance
that the balancing account needed to true-up annually for cumulative revenue-neutrality should run a small net balance, averaging sufficiently close to zero,
and probably reducing estimation errors by learning over time.
750. We rejected two alternatives sometimes discussed. Size class could be considered in size-specific feebate calculations under a single uniform pivot
point for all vehicles, but that would be more complex and less transparent. Or feebates could be customized for each automaker: that would of course be
manufacturer-neutral but at the expense of distorting customer choice between subclasses, removing a spur to competition between automakers, and creating a potential barrier to market entry.
751. Japan’s “Top Runner” system of improving light-vehicle efficiency (note 225, p. 45) also permits trading between classes.
752. This is one of feebates’ main advantages over the NAS/NRC 2001 proposal of merely grafting tradeable permits onto the system of CAFE standards
(unless the trading cap decreased over time).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
187
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Feebates
Unlike fuel taxation,
feebates directly
signal the value of
efficiency to the
buyer at the time
and place of choosing
the vehicle.
$10b/y,753 while customers would save ~$15b/y through 2020.754 Astute
manufacturers will also multiply their higher volumes times the higher
margins earned by superior value proposition and first-mover advantage.
Treating feebates
by size class also
nearly eliminates
the potential for
shifting customer
choice between
classes, and avoids
complaints about
interference with
freedom of choice.
A business buying a
powerful, highcapacity Class 2b
truck would pay no
more (and may pay
less) for a fuel-efficient version of that
vehicle with feebates
than for an inefficient
version in a system
like today’s, without
feebates—and that
business will also
pay far less for fuel.
Feebates would apply to all new light vehicles, rather than distorting the
market by discriminating between different types (e.g., CAFE exempts
the heaviest light vehicles, like Hummer, and the gas-guzzler tax applies
only to cars, leading makers to ensure that their least efficient vehicles are
legally classified as light trucks so they avoid the tax; otherwise a 15-mpg
Expedition would cost $3,000 extra). We see no reason to perpetuate in feebate structure the highly gamed historic distinction that CAFE regulations
now draw between cars and light trucks,755 nor between imported and
American-made vehicles; these simply distort the market.756 We haven’t
modeled size-class disaggregation, but, as Greene et al. (2004) found by
modeling up to 11 subclasses, it wouldn’t make much difference to our
findings.
Endorsed by the National Research Council’s 2001 study, feebates would
best be adopted federally for uniformity to both manufacturers and customers, and for ease of administration. In time they would supplant CAFE
standards by making them (and the gas-guzzler tax) irrelevant, though for
the time being those should remain to deter recidivism. But if the federal
government failed to act, many states are poised to step in, having already
considered such initiatives for more than a decade.
In 1990, for example, the California Legislature approved an early CO2based feebate, DRIVE+, by a 7:1 margin, though it was then pocket-vetoed
by Governor Deukmejian because one of the Big Three opposed it (another
was neutral; the third’s position is unknown). The concept enjoys 3:1
support in recent California private polling and is likely to return there
soon. A state feebate can easily piggyback onto existing new-vehicle sales,
excise, or use taxes. Massachusetts and Vermont have long considered
varying a 5% state tax rate within a range of 0–10%, based on efficiency
compared to other vehicles in each vehicle’s size class. Similar discussions
are underway in at least five other states.757 The vehicle’s sales sticker
could and should show how its sales tax is derived from its federally rated
mpg—though the Department of Transportation criticized that display of
753. Greene et al. (2004) found that under a $1,000 per 0.01 gpm feebate, manufacturers’ sales revenues would increase by $9.7 billion—the excess of
revenue gains (from selling more expensive vehicles) over revenue losses (from selling fewer vehicles).
754. The per capita savings are based on a calculation by the California Air Resources Board on the savings of California consumers under a nationwide,
carbon-based feebate system (Ashuckian et al. 2003, pp. 3–16).
755. This distinction would remain in CAFE rules unless, as we hope, NHTSA abolishes it as an arbitrary anachronism (all the more so with the rapid rise of
crossover vehicles). However, it would become irrelevant in step with CAFE itself as fleet efficiencies rapidly improved under the influence of feebates.
We also see no current reason to exempt any class of customers (e.g., farmers) from feebates, which would affect only new vehicle purchases, not existing
vehicle operations, and are meant to offer all kinds of customers more efficient choices of all kinds of road vehicles, but wouldn’t apply to vehicles meant
and licensed for almost exclusively off-road use, such as tractors.
756. The UAW asserts that this CAFE rule discourages automakers from moving production offshore. Our policy recommendations should considerably
increase the net benefit to automakers of sustaining and expanding U.S. production.
757. Totten & Settina 1993.
188
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Feebates
vital customer information in an unadjudicated advisory opinion that
effectively discouraged Maryland from enforcing its 1992 feebate law (and
deterred many other states from adopting feebates), on grounds that now
appear clearly unsound758 and readily avoidable.759
The Province of Ontario’s feebate system, operating since 1991, has had
predictably inconclusive results because it’s invisible to customers and
offers only a flat C$100 rebate, although the sliding-scale fee on gas-guzzlers can rise to C$4,400. However, it’s interesting because it was adopted
by a consensus of automakers, the Canadian Auto Workers, car dealers,
and environmentalists.760 Similar incentives are also gaining popularity in
Europe, and are gradually evolving from one-sided inefficiency taxes (all
stick, no carrot) to full feebates. Austria in 1992 raised its 32% new-vehicle
purchase tax to 37% for the least efficient models, but scaled the increase
down to zero for the most efficient. Efficiency-based fees are also in force
in Denmark, Sweden, and Germany. Diverse jurisdictions from Europe to
some Australian states base fees on vehicle weight.761 A full feebate likely
to take effect on 1 January 2005 in France is graduated in seven steps: it
would charge €3,500 for the least efficient one-third of new light vehicles
(such as large SUVs) while rebating up to €700 to the cleanest and most
efficient one-sixth. French automakers make mainly midsize family cars
(the half of sales that would have no net fee or rebate) or more efficient
ones, which would get rebates, so they quietly favor the proposal, which
would mainly crimp sales by their German and Japanese competitors.762
Implementation
Under a $1,000/0.01-gpm
feebate, automakers’
net sales would
increase by nearly
$10b/y, while customers
would save ~$15b/y
through 2020.
A full feebate likely
to take effect on
1 January 2005 in
France would charge
€3,500 for the least
efficient one-third of
new light vehicles
(such as large SUVs)
while rebating up to
€700 to the cleanest and
most efficient one-sixth.
French automakers
quietly favor
the proposal.
Because they work at both ends of the purchase decision—carrot and stick
combined—and affect what most buyers focus on intently (the net purchase price of the vehicle), feebates are strikingly effective. A $1,000/0.01gpm U.S. feebate “slope,” illustrated in Box 18 and compellingly analyzed
by Greene et al. (2004), would let new-car buyers behave as if they were
considering fuel use or savings over the entire 14-year nominal life of the
vehicle (no matter who owns it), rather than only for its first three years
as they do now. That is, this $1,000/0.01-gpm slope is societally efficient
758. Chanin 2003, especially pp. 747–753; Clinton 1999 reflects the constitutional principles of states’ rights and establishes a strong administrative presumption in favor of promptly issuing federal preemption waivers requested by states, except where preemption is clearly vital to the national interest. It cannot,
however, bind judicial determinations of the zone of federal preemption. Chanin (2003) compellingly argues that states need not even invoke CO2 regulation
(as California did with the 2002 Pavley Act, AB1493, which many other states may effectively adopt) to implement non-preempted feebates, provided their fee
is not so exorbitant as to leave buyers only a Hobson’s choice. (Her logic appears unaffected by the Supreme Court’s preemption decision of 28 April 2004 in
an air-quality case (AQMD 2004; U.S. Supreme Court 2004).
759. Absent federal leadership via uniform feebates, states could apparently adopt policies that reasonably influence (not mandate) vehicle choice without
restricting what makers may produce or customers buy, and by separating the required federal mpg label from any state feebate label calculating a feebate
from the federal mpg rating. Caselaw (Chanin 2003, pp. 752–754) doesn’t appear to hold that the “related to” clause in CAFE preemption stops states from
doing anything that might mention or influence customers’ efficiency or emission choices in any way (e.g., by the existing fourfold differences in state gasoline taxes). A different issue might arise in California under AB1493—that law prohibits the California Air Resources Board’s automotive CO2 regulations from
imposing vehicle taxes—but nothing stops the state legislature from doing so (it already does) or from basing taxes or feebates rationally on CO2 emissions,
weight, or even simply on federally rated mpg (Chanin 2003).
760. Morris 1994, “Ontario’s Automobile Feebates.” However, the consensus was under governmental duress.
761. Michaelis 1997.
762. Henley 2004. In presenting this key element of his National Health and Environment Plan, Prime Minister Raffarin noted that air pollution kills more than
30,000 French citizens a year, and 7–20% of cancers in France are believed to have an environmental cause. A senior UK advisor has suggested a 3–4×
increase in SUVs’ ~$300/y road tax (Power & Wrighton 2004).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
189
Implementation
763. For example, we
haven’t analyzed the potential to pay feebates not to
autobuyers but to automakers, thus leveraging the
roughly twofold markup
from bare manufacturing
cost to MSRP—though
making the price signal
opaque to the customer
and relying entirely on competitive forces to pass it
through to the retail level.
(Some manufacturers fear
the public may not trust
them to do this, creating ill
feeling.) Manufacturer
rebates have been done
advantageously with
smaller energy-saving
items, notably compact
fluorescent lamps (von
Weizsäcker, Lovins, &
Lovins 1997, p. 166).
Moreover, about a fourth of
the successful U.S. utility
incentive programs to promote electric efficiency in
the 1980s and 1990s
rewarded not just retail
buyers but also “trade
allies”: e.g., PG&E got a
bigger saving at a third the
cost by paying rebates not
to buyers of efficient refrigerators but to the appliance-store associate who
sold them. The chance to
earn this $50 bonus rather
than nothing (for selling an
inefficient unit) led the
stores to stock only efficient units; the rest were
sold elsewhere or, one
hopes, not made. Another
option: pay dealers’ carrying charge for very efficient
cars—enough incentive
to swing a market (as B.C.
Hydro did to flip the market
for big electric motors in
just three years [von
Weizsäcker, Lovins, &
Lovins 1997]). States that
wanted to save oil faster
could experiment with such
high-leverage dealer or
salesperson incentives
even under a uniform
federal feebate.
190
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Feebates
because it arbitrages the difference in the discount rates used by new-car
buyers and by society. Our model indicates (Fig. 36) that a feebate with
this slope would raise the 2025 new-sales share of Conventional Wisdom
vehicles from 19% to 44% for cars and from 45% to 56% for light trucks.
Because of their later arrival into their market and their slower retooling
(Fig. 37a), State of the Art vehicles’ new-sales share would rise from 20% to
only 45% for cars, and from 18% to only 43% for light trucks. Because SOA
vehicles are more efficient, they’d be 11% of the 2025 stock but provide
34% of oil savings from non-EIA vehicles.
Many, including ourselves, would consider it economically efficient to
double the feebate slope to $2,000/0.01 gpm to begin reflecting the public
value of reduced oil use and a revitalized light-vehicle sector. (This is
economically equivalent, at our 5%/y real discount rate over 14 years,
to valuing saved gasoline at an extra 26¢/gal and saved short-run marginal crude oil at an extra $3.6/bbl—far below the externality estimates
on pp. 21–22.) The resulting fees and rebates would then approach the
~$4,000–5,000 rebates now commonly paid by manufacturers to customers to promote the sale of large SUVs—but, in complete contrast to
those incentives, feebates wouldn’t reduce the automakers’ profit margins. A $2,000/0.01 gpm feebate would increase State of the Art vehicles’
market share by another five percentage points, although SOA share in
the on-the-road fleet rises by only half that much because of the early
years’ production bottleneck. Despite the persistent production constraint,
the higher feebate saves an extra ~0.2 Mbbl/d in 2025 (almost all from
increased State of the Art sales). The accelerated State of the Art introduction would save even more oil thereafter—by 2035, an extra 1.23 Mbbl/d.
We have not evaluated even higher feebate slopes, though our downloadable model can easily be used to do this.
Feebates should be considered not just for light vehicles but also for
medium and heavy trucks, perhaps for other vehicles such as buses and
trains, and conceivably for airplanes (for which we made a different suggestion on p. 156, tailored to that industry’s unusual business conditions).
Broadly speaking, anything that moves and uses oil would be a potential
candidate for this powerful, flexible, simple, market-oriented, and versatile policy instrument and its variants.763 Indeed, feebates are so powerful
that in time, they could achieve all the benefits of our entire policy package (except retooling incentives). But to increase the speed and confidence
of the shift to advanced technology vehicles, to capture further benefits,
and to ensure a balanced and diversified group of mutually reinforcing
instruments, we also suggest a wider portfolio, including the other policies described next.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
Low-income scrap-and-replace program
Welfare reform has been the subject of great interest over the past decade.
The root causes of poverty in America have been identified as a combination of lack of capital, lack of skills, failure to form families,764 lack of reliable transportation,765 and low-wage earners’ need for certain welfare benefits not typically provided by low-wage jobs, such as health care.766 Much
effort has been put into strategies that address these issues, specifically
programs that help welfare recipients transition to the working world.767
A major question is how to keep them working and help keep them from
reverting to poverty. Our focus here is on the impact of improving timely,
reliable transportation for low-income Americans, going beyond the laudable existing public transport initiatives to address issues such as transport
for those who work outside peak travel times of public transit, and for
those living in rural areas not served by public transit. There is a growing
consensus that limited mobility is an important “missing link” in a comprehensive strategy for reducing and ultimately eliminating poverty.
Our second policy initiative therefore links and thus helps to solve three
distinct but related problems: lower-income Americans’ limited personal
mobility and disproportionately burdensome fuel purchases (both part of
the “poverty trap”), their inability both to gain from and to contribute to
the prompt benefits of transforming automaking, and the slow (~14-year)
spontaneous turnover of the nation’s light-vehicle stock, which retards
the adoption and hence the development of more efficient models.
We outline two alternative ways to address these three problems. Both
proposed mechanisms involve best-practice768 and preferably nationwide769 scrappage of qualified inefficient cars. They also create a profitable
new million-vehicle-a-year market to replace the scrapped ones; increase
oil savings and pollution prevention; and improve equity and social
welfare. The two mechanisms differ only in which low-income used-car
buyer segment they help—the most marginal used car buyers, who can
barely afford a used car, or the least marginal buyers, who already buy
first-generation used cars. We recommend that a neutral expert body
like the U.S. Government Accountability Office (GAO) determine which
mechanism and design details would produce greater social welfare benefits, and recommend a policy, which may well include both mechanisms,
for pilot tests to support nationwide rollout.
766. Fraker et al. 2004, p. 16.
767. Sawhill 1999.
Implementation
Low-income
Americans pay as
much for driving as
for food. But about
$0–3 per day extra
can provide a very
efficient and reliable
new car plus its fuel,
while a clunker gets
scrapped. Affordable
personal mobility
may be the missing
link in reducing
welfare dependence
while opening a
new million-vehiclea-year market.
We aim to create the
same revolution in
affordable personal
mobility as was
achieved in affordable home-ownership
after World War II.
If that mobility is
superefficient,
a low-income household can afford not
only to get but also
to drive a reliable
vehicle, gaining
the mobility that is
the key to America’s
opportunity.
764. ISP/ASPE/HHS 1983;
NCSL, undated.
765. “According to the Community Transportation Association of America, JOBS program
studies have concluded that the lack of affordable transportation presents a barrier even
more serious than the lack of child care to prospective JOBS clients.” Kaplan 1997; Cotterill
& Franklin 1995.
768. Kallen et al. 1994.
769. Horowitz (2001, p. 121) shows that “… it is far more cost-effective to attempt to transform a national market through long-term coordinated coast-tocoast efforts that permit market preferences to evolve and mature, than it is to temporarily manipulate local markets through piecemeal programs that are
highly variable from place to place and from year to year. In short, persistent efforts to educate producers and consumers and inform them of energy efficiency benefits appear to be more capable of building sustainable sales volumes and market shares than the alternative of financial subsidies.” (Note 29 in
IEA 2003b). We recommend staging from regional pilot programs to national rollout.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
191
Implementation
770. IEA 2003b.
771. Raphael & Stoll 2001.
This is basically because in
most U.S. metropolitan
areas, “one can commute
greater distances in shorter
time periods and, holding
distance constant, reach a
fuller set of potential work
locations using a privatelyowned car rather than public transit” (which also may
not match the irregular
hours of many low-income
workers’ jobs). Other benefits include greater ease of
getting and holding a job
(because one can get to it
reliably), more family time
and stable homes (less time
on often slow public transportation), access to a
wider choice of food and
other goods at more competitive retail prices than
the urban core offers, and
less or no exposure to
volatile gasoline prices.
Of course, the quoted
statement doesn’t mean
that only black and Latino/a
families make up the lowest
income quintile; most lowincome rural Americans are
white. However, black families accounted for 16% of
the lowest-income quintile
in 2002 (“White and Other”
accounted for the remaining 84%), and for decreasing percentages of the
higher income quintiles—
14%, 11%, 9%, and 6%
(BLS, undated).
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Scrap-and-replace program
Either way, the goal is clear and compelling: to create the same revolution in affordable personal mobility as was achieved in affordable homeownership after World War II—applying to all U.S. working families an
expanded version of Henry Ford’s vision that his cars should be affordable to the workers who built them. But our goal goes further: to make
that mobility superefficient, so that a low-income household can afford
not only to get but also to drive a reliable vehicle, gaining the mobility
that is the key to America’s opportunity.
Affordable and superefficient personal mobility, especially if offered in
many states or nationwide,770 would bring life-changing benefits to lowincome Americans, especially in areas poorly served by public transit:
for example, “raising minority car-ownership rates to the car ownership
rate of whites would narrow the black-white employment rate differential
by 45 percent and the Latino[/a]-white employment differential by 17
percent.” 771 The vehicles driven by low-income citizens are not only slightly less efficient,772 but also tend to be the least reliable and often the most
polluting and least safe in the fleet. Correcting these conditions and making personal mobility affordable is more valuable than the gasoline savings, which for a nominal million-vehicle-a-year program would accumulate over a car’s lifetime to ~0.2 Mbbl/d with Conventional Wisdom or 0.4
Mbbl/d with State of the Art compact cars.773
An instructive analogy comes from America’s historic national commitment to make home ownership more affordable.774 Many can now afford a
home 775—but not driving between home, job, and retail stores. That’s costly at best, and when gasoline prices soar, it can become prohibitive,776
making some families choose between feeding the children and fueling
the car, without which there’s no paycheck. For the average low-income
household, mobility is normally as costly as food.777 In cities with long
commutes, mobility now costs more than housing.778 Yet without credit, a
marginally better car is a stretch and a new car is an impossible dream.779
772. Khazzoom (2000a, p.26) notes that 1993 U.S. households with average incomes ≤$15k vs >$50k had average vehicle efficiency of 19.8 vs 20.1 mpg.
No newer data are available, although qualitatively, the 2001 partial Department of Transportation data appear comparable. Driving was 49% lower in the 1993
lower-income households (14,109 vs. 27,740 miles).
773. Assuming savings from annual replacement of 1 million 23-mpg MY1985 cars with 37.5- and 90-mpg cars over 2010–24 (14-year average lifetime), saves
0.2–0.4 Mbbl/d.
774. Andrew 2004. Home ownership rose from 64% in 1994 to nearly 69% in 2003; among African-Americans, from 42% to 48%; among Latino/as, from 41% to 46%.
775. At least to finance the home if not to pay the utilities—efficient building and appliance design typically gets too little attention.
776. Ball 2004. The squeeze has reportedly reduced mid-2004 revenues at Wal-Mart and Target.
777. Tan 2000, pp. 29–35. In 2002, before the run-up in gasoline prices, the average low-income (bottom-quintile) U.S. household devoted 35% of its $19,061
annual income to housing, 17% to mobility (chiefly by car), 17% to nutrition, 9% to utilities, fuel, and public services, and 7% to health care. In contrast, the
average household’s $57,835 income was split 32%, 20%, 13%, 6%, and 5%, respectively.
778. Ball 2004. In sprawling Tampa, where 75% of jobs are over ten miles from the city center, transportation costs more than housing.
779. Three-fourths of U.S. families earning less than $15,000 a year own cars; otherwise, lacking effective public transportation, they’d be largely immobilized.
But investment in automotive efficiency or reliability is beyond reach. In that paycheck-to-paycheck world, cars tend to be fourth- to sixth-hand clunkers
bought for cash, tinkered with and coaxed into running for just another month.
192
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Scrap-and-replace program
Implementation
Raising minority car-ownership rates to the car ownership rate of
Fortunately, substituting clean,
whites would narrow the black-white employment rate differential by 45
efficient cars for dirty and often
percent and the Latino[/a]-white employment differential by 17 percent.
inefficient ones benefits everyone.
Prematurely retiring “super-emitters” 780 would disproportionately save oil, save money for those most in
need, and improve urban air and public health.781 Specifically targeting
such cars is a cost-effective way to reduce pollution, and even the broader
scrappage we propose is still worthwhile for society 782 So what are the two
ways we suggest to scrap qualified inefficient cars, save oil, and provide
780. I.e., vehicles emitting
efficient, reliable mobility to non-creditworthy lower-income citizens?
Our first proposed mechanism combines scrappage of qualified cars and
financing of new cars with high-volume procurement. A federal agency
such as the General Services Administration (GSA), which now buys most
federal government vehicles, would competitively procure, at a fair unit
margin to the winning automaker, additional but highly efficient new
vehicles. It would then lease these at federal rates to qualified non-creditworthy low-income citizens, possibly via competing for-profit firms.
The bulk-bought vehicles must correctly express aggregated customer
preferences for reliable, efficient mobility and for other desired vehicle
attributes.783 For every efficient new car bought and leased, slightly less
than one older, less efficient, more polluting vehicle would be scrapped.
For maximum flexibility, these two transactions need not necessarily
involve the same driver or household, although such linkage would be
simple and worth considering. The lease would be very inexpensive
because of the large-scale procurement, especially since the GSA would
be buying one very efficient model. The program can create five other
benefits, accruing to various parties but all monetizable: future saved fuel,
two emissions credits, cheaper insurance bought in bulk, and the lessees’
avoided purchases of replacement clunkers. Low-income lessees, who
suffer grievously when gasoline prices spike, would benefit further from
the program’s bulk-buying price-hedged (constant-price) gasoline,784
in much the way that many large private fleets now buy their fuel.
many times the average
pollutant levels per mile.
A recent study by major
car and oil companies
found from roadside realtime emission measurements in Denver that even
as newer, cleaner vehicles
emitted less of the smogforming pollutants, the
fraction emitted by the dirtiest tenth of the vehicles
held steady or rose: by
2003 it accounted for 69%
of the fleet’s CO, 75% of HC,
and 54% of NOx. (WBCSD
2004, p. 100). Similar issues
are far more acute in most
giant cities in developing
countries, where similar
remedies would be worth
considering—not just for
cars but also for, e.g., the
dramatically pollutionreducing two-stroke scooter retrofits developed at
Colorado State University:
a $200 retrofit can cut
CO emissions by ~1 ton/y
and cut fuel use 32% (M.
Defoort [Engines and
Energy Conversion
Laboratory, CSU, Ft. Collins
CO], personal communication, 2 August 2004).
781. Now that U.S. new-vehicle efficiency has been nearly flat for two decades, the opportunity to save a great deal of fuel by scrapping the oldest vehicles
is largely gone; but the least efficient vehicles are also often the heaviest, most aggressive, and least compatible with their lighter roadmates. Where this is
true, scrapping them would improve public safety (and slightly increase recovery of currently scarce scrap steel).
782. Not all scrappage is worthwhile. We tested Kelley Blue Book prices and EPA efficiency data to check the “scrappage resource” supply curves for some
popular models, relating their age to the $/bbl cost of saving oil by scrapping them. Most of the inherently inefficient heavy models are relatively new and
valuable, making their CSE relatively high. Scrapping older cars is often a more cost-effective oil-saver, but the amount saved is smaller.
783. See notes 812–813 below for lessons from other kinds of aggregation-and-procurement programs. Microcredit experience suggests that the default rate
may be surprisingly low, and the collateral is excellent.
784. Low-income customers would especially benefit from constant-price gasoline. The posted price at the pump would appear to vary, but the lessee’s
special credit card would be charged the constant price. (Car-sharing firms like FlexCar, whose hourly fees include fuel, use special credit cards that work
only with a specific vehicle and are electronically cancelled in the event of fraud or default.) The hedging could, for example, piggyback off the Defense
Energy Support Center (p. 87) hedge purchases for all DoD gasoline supply, which like other military fuels is provided at constant prices known far in
advance. Indeed, the new car’s high efficiency could motivate still another innovation: leasing not just a car but a mobility service solution where lessor and
lessee share the benefit of the fuel savings. This alignment of provider with customer interests illustrates the powerful “solutions economy” business model
(pp. 135, 196) described by Hawken, Lovins, & Lovins 1999, Ch. 7, and in the forthcoming book The Solutions Economy by J.P. Womack & D.T. Jones.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
193
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Scrap-and-replace program
The analysis in Technical Annex,
Ch. 22, finds that low-income customers could lease an efficient new
compact or midsize Conventional
Wisdom car—safe, clean, comfortable, insured, reliable, warranted, with a five-year extended service
plan—for an incremental cost of ~$3/day without feebates, or ~$2.3/day
with feebates. Starting around 2010, State of the Art vehicles with feebates
could reduce that incremental cost to about 15¢ a day. For early adopters,
the “Golden Carrot” aggregated-procurement incentives described on
pp. 199–200 could even cut that to zero—or perhaps even less, earning a
return to help finance participation for the neediest.
Many can now afford a home—but not driving between home, job, and
retail stores. For the average low-income household,
mobility is normally as costly as food. In cities with long commutes,
mobility now costs more than housing.
785. Moral hazard would be
minimized via ensuring certain disclosure requirements relating to an individual lender’s historical lower
threshold for making credit
available, and via the ability
to compare confidentially
the thresholds across multiple lenders.
Our second proposed mechanism includes the same scrappage program,
but doesn’t use government procurement, and has a different financing
mechanism. It would provide affordable mobility in the form of efficient
new cars, but this time to customers who buy, not lease, and whose credit record marginally disqualifies them from buying a new car. This program would similarly retain the demand-supply balance in the used car
market, and would therefore raise quality for all used-car buyers rather
than disproportionately for the lowest-income buyers. The financing
mechanism would simply guarantee reimbursement to current auto
lenders for incremental defaults on loans made to the marginal next-inline new-car borrower category, and would therefore almost exclusively
engage existing market mechanisms and financial institutions. The cost
would largely consist of reimbursing the incremental defaults.785 This is
analogous to the student loan program developed to ensure financing for
this roughly equally risky customer segment.786 It could involve a separate office with sole responsibility for reimbursing the incremental
defaults, but there would be no need for a “designated agent” such as
car dealers or car rental companies.787 Regardless of who originates the
retail loans and whose cars they finance, a “Carrie Mae” institution could
provide a reimbursement guarantee for financing efficient new cars into
a new low-income market while relying on today’s financial services
infrastructure for screening and execution. Existing financial services
companies (FSCs) would compete for a piece of the action,788 given an
786. The Federal Family
Education Loan Program
Program (FFELP), and to
some extent Sallie Mae
(SLMA, the Student Loan
Marketing Association).
Sallie Mae was established
by Act of Congress in
1972 as a GovernmentSponsored Enterprise,
floated on the NYSE in
1984, privatized in 1997,
and renamed in 2002 (Sallie
Mae, undated). When the
Education Act was passed
in 1965, FFELP was set up
to cover any defaults on
student loans that were
issued by banks, but by the
early 1970s, this repayment
provision for delinquent
loans proved too slow and
inadequate, and the banks
underserved the demand
because the lag between declaration of default and government repayment crimped the lenders’ liquidity. Sallie Mae fixed this problem by buying the student loans very shortly after issuance. This increased both liquidity and loan volume: relieved of default risk, each bank could fully satisfy loan demand from
the next student cohort.
787. Instead, dealers, who have traditionally originated loans on-the-spot mainly for their parent automakers’ financing arm or third parties, would compete
with newer entrants, such as the financial service companies (FSCs) that lend to many segments of the consumer market. FSCs also finance cars, and have
recently become very efficient at doing so by mailing directly to qualifying buyers a voucher or check spendable at any dealership. FSCs do this today for
relatively safe customers, mailing vouchers just like preapproved credit-card offers to those with FICo scores above ~660 (Fair Isaac’s Company, Inc.
scores—a standard consumer-finance rating system analogous to Moody’s or S&P, with a maximum possible score of 800). Under our proposal, the new-car
voucher simply has far better collateral (note 792).
788. FSCs already know the sub-prime auto market very well, and would likely be eager to expand their servicing via deepening, if a reimbursement guarantee were to be offered. Carrie Mae would define qualifying levels of new-car efficiency, as well as mechanisms for the FSC and Carrie Mae itself to certify
scrappage.
194
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Scrap-and-replace program
attractive business model.789 It would be essential to clearly define under
what circumstances the default reimbursement guarantee kicks in, so that
the risk profiles of these loans are completely defined by a qualification
mechanism that specifies who qualifies and for how much credit.790
The scrappage arrangement is identical for both mechanisms. Carrie Mae
would define qualifying levels of inefficiency and mechanisms for the
buyer or lessor to certify scrappage. “Cash for Clunkers” programs that
pay customers or bounty-hunters to scrap old-but-still-driven cars have
been implemented or considered in at least five states (AZ, CA, DE, IL,
ME),791 revealing important program-design lessons.792 Vehicle scrappage
must target only vehicles that would otherwise actually drive enough
miles to make it worthwhile.793 And to avoid driving up prices for parts
for collectors’ vintage vehicles, either the public should be allowed to
scavenge them before scrappage (with notice to collectors’ groups when a
noteworthy vehicle is received), or collectable vehicles should be excluded from the program. In any event, new-car procurement could be linked
with the “Golden Carrot” incentive described on pp. 199–200.)
789. The expected loan risk in this used-car segment is about the same as that of the student loans Sallie Mae buys today,
so with risk and cost guaranteed via the reimbursement policy, and the lender’s net loan-to-value ratio well below one,
interest should be high among FSCs. Sallie Mae loan clients (students) have ~20–40% default rates; sub-prime auto borrowers, an estimated ~20–35%. Although default rates are proprietary, one deep sub-prime auto lender—Texas-based
Drivetime, www.drivetime.com—reportedly charges well over 20%/y for its lowest-rated used-car buyers, who probably
FICo-score well below 550. This implies expected default rates around 25–40%, so for the riskiest new-car buyers, ~20–35%
seems reasonable. However, unlike an uncollateralized student loan, the FSC’s loan is secured by the new car, which is
easily traced, can be repossessed (wireless transponders are cheap), and is more valuable because of its high efficiency
and modernity. The guarantor’s cost is thus essentially the transaction cost of depreciation, repossession and resale, with
significant principal remaining. Default rates should be further reduced because nearly 100% financing (no down payment,
just normal finance costs) improves the borrower’s strained cashflow; most repairs are free under warranty; and the
increased ability to get to work reliably reduces the risk of losing one’s job if the car breaks down. The lower-income borrower’s new ability to build a sound credit history and equity in the new car—a far more valuable asset than the previous
clunker—is an added incentive. Some sub-prime lenders even install radio-controlled devices that disable a car on its next
start if the loan is in default after several days’ warning, but this might prove inappropriate or unnecessary.
Implementation
Low-income
customers could
lease an efficient
new compact or
midsize Conventional
Wisdom car—
safe, clean, comfortable, insured,
reliable, warranted,
with a five-year
extended service
plan—for an incremental cost of
~$3/day without
feebates, or ~$2.3/day
with feebates.
Starting around 2010,
State of the Art
vehicles with feebates could reduce
that incremental cost
to about 15¢ a day.
“Cash for Clunkers”
programs that pay
customers or bountyhunters to scrap
old-but-still-driven
cars have been
implemented
or considered
in at least five states.
790. For example, if an FSC’s historical baseline cut-off FICo score were 660 and if, after seeing the terms offered by the
reimbursement guarantee, the FSC were induced to deepen its lending cutoff point to 640, the reimbursement would pay
for any defaults within this FICo range that were incremental to the FSCs documented historical default rate at the 660
level. So if defaults went from 20% at the 660 level to 23% for the 640-to-660 borrowers, the guarantee would reimburse 3
percentage points of those guarantees, or 13% of all defaults for this segment. With this guarantee, the FSC could treat the
640-to-660 borrower segment on an equal footing with the 660-level borrowers to whom the FSC already lends.
791. And about a dozen other countries, mostly paying ~$500–1,500 and meant to reduce local air pollution (IEA 2001, pp.
87–89). The most ambitious effort, in Italy in 1997, scrapped 1.15 million light vehicles or ~4% of the national fleet in a year.
Payment was scaled to engine displacement, and was conditional on buying a new car—conveniently for firms like Fiat—
but sales then dropped, predictably, by about the same amount.
792. A decade-old but still useful guide is Kallen et al. 1994.
793. E.g., by strictly enforced requirements such as valid registration for at least one year prior to scrapping and demonstrating roadworthiness by driving vehicles to the dealership, then ensuring that vehicles turned in are scrapped so that
they do not re-appear. Empirical evidence (Dill 2004) suggests that scrappage incentives tend to attract lower-income
households that drive their clunkers more, rather than higher-income households seeking to dispose of a surplus vehicle.
However, low-income households with the highest driving levels are unlikely to be attracted by a $500 scrappage bonus
because that’s not enough to buy a replacement clunker. About three-fourths of vehicles scrapped in Bay Area programs
would otherwise have been driven, as program designers assumed, for about three more years (Dill 2004). Some programs
specifically target the most polluting cars (World Bank 2002), or those that have just failed smog tests. Some, like British
Columbia’s Scrap-It, offer a variety of compensation options including a free transit pass (which most participants choose
[Clean Air Initiative Asia, undated]).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
195
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Scrap-and-replace program
Automakers would
sell a million additional vehicles a year
into this new market
and would make a
profit on every one
of them. That offtake
would also lower
the risk of developing
new fuel-efficient
platforms.
So far, most scrappage programs have been on a sufficiently modest scale
not to dry up the clunker supply and make older used cars unaffordable
to those who most need their mobility. However, a successful large-scale
scrappage effort could in principle have a slightly regressive effect,799
chiefly affecting minorities and low-income rural residents.800 This would
be offset in part by the decline in used-car prices when the leased cars are
resold into the used car market, since pre-leased cars typically resell for
less than pre-owned private cars. Further, the estimated ~1:1.2 ratio of
inefficient old cars scrapped to efficient new cars financed (to be adjusted
in the light of experience) avoids regressivity at the low end of the usedcar market by not scrapping one additional used car for every eleventh
20: More antidotes to regressivity
The combination of the suggested low-income
lease-and-scrap program plus feebates (for
which all new-car buyers are eligible) should
eliminate the potential for burdening lowincome drivers, and on the contrary should
much improve their access to affordable personal mobility. However, if needed, further
options are available.
Our analysis hasn’t fully counted all ways in
which careful financial engineering can cut risks
and hence costs. (By such methods, one firm
has systematically cut public-sector and university housing finance costs by more than half.)794
Car-sharing programs795 in the U.S. and Europe
typically cut annual driving by ~30–70% without
loss of convenience or mobility. Some services,
like FlexCar,796 claim to take six cars off the road
for every car they field; the hourly lease fee
includes premium insurance, fuel, reserved
parking, and maintenance. ZIPcar claims 7–10×
reductions, implying that some infrequent drivers
give up their cars in favor of cheaper occasional
rentals. A more fully integrated offering like the
58,000-member, 1,750-vehicle, 1,000-site Mobility
Car Sharing Switzerland797 delivers superior
mobility at lower cost with no self-owned car.
Such concepts may hold special promise for
low-income communities to cut costs.798
794. UniDev LLC (www.unidevllc.com), for a good-sized California
project, recently cut the annual income required to qualify for a
$338,000 house from $106,000 to $43,000 and for a 1,000-ft2 two-bedroom apartment from $81,000 to $39,000.
795. WBCSD 2004, pp. 139–141. The European and U.S. car-sharing news sites are respectively www.carsharing.org and www.carsharing.net; the
World CarShare Consortium is at http://ecoplan.org/carshare/cs_index.htm. U.S. car-sharing typically saves money if you drive less than ~7,500 mi/y,
and is available in more than three dozen cities. We aren’t aware of a car-sharing service designed specifically for low-income U.S. communities, but
suspect it could offer some attractions.
796. See www.flexcar.com/vision/impact.asp.
797. This service, www.mobility.ch, is now available to ~60% of the population of Switzerland. It combines in a single fee a free or discounted pass
for public transportation and a short-term service to drive the vehicle of your choice (in some cities, dropped off to and picked up from your location
by a bicycle courier). It could be extended to include more services, such as a cashless backup taxi service, a travel agency, and a broadband
Internet service provider (for videoconferencing and, soon, virtual presence). Such business models work because not everyone needs a car at
once; the average U.S. private car has an asset utilization of only 4%, so it stands idle 96% of the time.
798. A possibly apocryphal, but financially plausible, story relates that several hundred residents of a distressed Massachusetts town sold their cars,
pooled the proceeds into a nonprofit group, and bulk-bought a fleet of identical, efficient Honda cars—enough to get the dealer price, and thanks to
the nonprofit ownership, tax-free. Every two years, they sold the fleet and bought a new one, thus staying in warranty and remaining a qualified dealer. They also bulk-bought insurance, fuel, and a two-person maintenance staff (formerly employed seasonally to keep up the town’s snowplows). Carsharing—cars were booked and checked out like a library book—increased asset utilization, providing more mobility from fewer cars with lower
total cost. The alleged net annual savings totaled about $5,000 per household—a huge boost, equivalent to one-fourth of total income for the bottom
quintile of American households.
196
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Scrap-and-replace program
Implementation
new car bought.801 This ratio also aids market entry by households that
have not previously owned a car—a fraction that’s been declining but
that, in 1999, still included 27% of households below the poverty line.802
(Box 20 mentions other remedies for regressivity.) Since cars would qualify for scrappage based on the “Reasonably Reliable Test,” 803 we conclude
there is little reason to sunset the scrappage policy (hence also the financing mechanism) before 2025. If saturation of the target segment(s) ever
became problematic, that would be a nice problem to have and is best
dealt with at the time. As more efficient new cars make old ones relatively
less efficient, the societal value of accelerating their scrappage rises.
Under either mechanism, all Americans, especially those with lower
incomes, would enjoy the same or greater personal mobility, and average
real used-car prices wouldn’t rise (because the scrap/replace ratio is less
than one). Low-wage earners would see lower operating costs and higher-quality, more reliable personal mobility at a cost, compared with their
current arrangements, ranging from a few dollars a day to zero or even
slightly negative. Since the new cars meet far stricter emission standards
than the scrapped clunkers, air pollution should decrease. High fuel efficiency would reduce CO2 emissions too. Congestion probably wouldn’t
increase appreciably, because most of the new trips by previous non-carowners would be in the opposite direction to inbound suburbanites’ commutes, and many would also occur at different times. And if the program
works as intended, automakers would sell a million additional vehicles a
year into this new market and would make a profit on every one of them.
That offtake would also lower the risk of developing new fuel-efficient
platforms.804
Smart government fleet procurement
The 2002 federal vehicle fleet, both civilian and nontactical military,
contained more than 470,000 light vehicles and 21,000 heavy trucks.
When state and municipal fleets are included, a total of nearly four million
vehicles, over half of them light vehicles, are a part of one government
799. A $1,000/0.01 gpm feebate rate (Greene et al. 2004) would probably reduce demand for cars by ~0.5%, reducing lightvehicle replacement rates by ~100,000 cars/y in 2025. But over the medium-to-long term, this reduces the supply of cars to
the secondary market, increasing used-car prices by a couple of percent: e.g., a price elasticity of demand of –1.0 implies
that scrapping 1 million cars would raise the average used-vehicle price by $198 (2000 $) in 2005, based on the July 2003
average used-vehicle price of $9,092. See also note 801 and Box 20.
800. Raphael & Stoll 2001; Schachter, Jensen, & Cornwell 1998; Stommes & Brown 2002.
801. The twelfth car offsets the regressivity that might be introduced by feebates. For economic efficiency and Pareto optimality (making someone better off but nobody worse off), we propose decoupling the purchase and scrappage transactions. This would indeed permit focusing scrappage on the least-efficient vehicles, further increasing oil (and perhaps pollution) savings, but our model doesn’t assume this.
802. BTS 2001, Fig. 1. Fig. 2 shows 18% of 1999 households were without cars in the central city but only 4% in rural areas.
Governments buy
billions of dollars’
worth of light
vehicles every year.
They can help mold
the market by shopping as intelligently
as a demanding
private purchaser.
The federal
government alone
spent $1 billion
in 2002 to buy and
lease light vehicles;
all governments,
billions of dollars.
803. Kallen et al. 1994.
804. Initially, the lease-and-scrap program adopts Conventional Wisdom vehicles because of their much lower cost and
early availability; later, the program could shift to State of the Art cars as they became available. This will increase the rate
of saving oil and may reduce the incremental cost to lessees, especially with feebates.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
197
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Fleet procurement
The federal fleet
alone in 2002 (light,
medium, and heavy
vehicles) drove
five billion miles
and used a third of a
billion gallons of fuel.
fleet or another.805 The federal government alone spent $1 billion in 2002
to buy and lease light vehicles; 806 all governments, billions of dollars.
Taxpayers would be better served if the several hundred thousand annual
vehicle purchases had efficiency and operational strings attached.
805. FHWA 2002; GSA 2003.
806. GSA 2003.
807. Executive Office for
Administration and Finance
(Commonwealth of Massachusetts) 2003.
We recommend that federal and state agencies be promptly required to
purchase American-made vehicles from the 10% most efficient in their
class, subject to operational requirements. This extends the policy of
Massachusetts, whose Republican Governor, Mitt Romney, requires purchase of only the most efficient and best-lifecycle-value vehicles, all with
ultra-low emissions and no less than 20 mpg city rating (EPA adjusted).
His policy also allows purchase of four-wheel-drive vehicles only when
“absolutely necessary for emergency or off-road response” (not just helping non-emergency state personnel get to work in inclement weather),
and light trucks only where justified by need (such as pickups for highway cleanup); otherwise more efficient platforms must be substituted.807
Comparable smart procurement is appropriate for light, medium, and
heavy vehicles. Since the federal fleet alone in 2002 (light, medium, and
heavy vehicles) drove five billion miles and used a third of a billion gallons of fuel,808 the fuel savings can be considerable. Citizens will have
the educational value of seeing more of the most efficient models on the
road and the satisfaction of seeing their tax dollars better spent. The shift
of market share to makers of the most efficient platforms will encourage
the other automakers to develop such platforms. Lifecycle-cost acquisition policies should become universal, and would be strongly reinforced
by feebates.
Currently, the top 10% of models within each size class (or subclass if so
restricted by operational needs) would include all hybrids and a smattering of other efficient models. To the manufacturer, hybrids today cost more
per saved gallon (2004 Prius ~$1.75/gal, Civic Hybrid ~$2.13/gal)809 than
they will in ~2007 as powertrains’ marginal cost falls by half. For example,
Accord Hybrid and Camry Hybrid should then both have a CSE around
80¢/gal. But that’s still well above our State of the Art ~2010 projection of
56¢/gal, which will itself be a moving target. And even State of the Art
vehicles, though very worthwhile, have a higher Cost of Saved Energy than
the far less efficient but 12¢/gal Conventional Wisdom vehicles. The public
sector should lead the market shift by backing the highest-saving cost-effective (SOA) vehicles, and until those are available, modern hybrids. Many
but not all private purchasers (Fig. 36) will choose to do the same, especially when even minimal ($1,000/0.01 gpm) feebates shift their fuel-saving
time horizon from 3 to 14 years.
808. GSA 2003.
809. These very rough estimates compare a 2004 Prius with a $4,000 (2003 $) incremental retail cost—the actual incremental retail price is considerably
lower—with a 36-mpg 2004 Toyota Echo (there’s no exactly comparable model because Prius is a “ground-up” new platform). We compare the 2004 Civic
Hybrid (5-speed manual, 48 mpg) with a 5-speed 2004 Civic nonhybrid (34 mpg) and use the actual $2,560 incremental price (2003 $).
198
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
“Golden Carrots” and technology procurement
Beyond routine government procurement, there’s a wider toolkit of
“technology procurement” for pulling new technologies into the market
faster. Many variants are in use in many countries. We have evaluated
which variants could fit U.S. light-vehicle markets. Box 21 lists some of
them, and suggests ample and flexible scope for technology procurement
with or without a government role. The focus here is not on the individual
technologies that go into a vehicle, but the whole vehicle as a system.810
Business risk-takers are best incentivized by real carrots, not just by “sticks
painted orange.”810a A particularly juicy flavor could use existing government procurement programs, often, but not necessarily, civilian, to pull
better technologies into production by guaranteeing to buy a certain number of them, for a certain number of years, at a certain minimum price,
if they meet certain specifications substantially more advanced than anything now available. The aim is thus not just to buy a widget, but to make
much better, even sooner, the widgets that one is buying anyway.
Such governmental “Golden Carrot” programs were pioneered by Hans
Nilsson and his colleagues in Sweden, whose Agency for Public Management (Department for energy efficiency [Kansliet] at NUTEK), among
other roles, aggregates all public-sector procurement, giving it strong
market clout. During 1990–2000, “Golden Carrots” were offered for
32 new energy-saving products in six major categories serving Swedish
government buyers and residential, commercial, and industrial customers.
Some of those products created new export markets. Various targeted
procurement designs were tested.811 Naturally, each sector turned out to
have different success factors, and success varied between projects. But
overall, the key to success was the vigor and strict neutrality of the procurement process, with no preference for any firms, countries, or designs.
We suggest procurement led by the federal General Services Administration and large states, to offer substantial, high-confidence markets to
early providers of advanced technology vehicles. (If aggregation is difficult because of state- or regional-level dealer-relationship arrangements,
it may be possible to do “virtual aggregation” unbundled from actual
procurement, with appropriate manufacturer netbacks to dealers.)
In addition, we suggest dangling an even more valuable kind of carrot
for which automakers would stretch even higher.
810. Technology procurement policies with government involvement often don’t involve the government as the primary
buyer. In cases when the government is the buyer, only in rare circumstances (e.g., military) does it guarantee sales numbers, price, and/or a time period for sales. The buyer could as well be a cooperative of private entities. Independent of
who’s buying and any level of guarantees, there is normally a coordinator of the whole exercise, and that coordinator is
often but not necessarily a government employee. In the absence of guarantees, the coordinator’s primary role is to aid
market development via a carefully planned set of steps detailed elsewhere; see Olerup 2001.
Implementation
Governments and
large fleet owners
buy hundreds of
thousands of vehicles
every year.
That purchasing
power can pull
very efficient vehicles
into the market
much earlier.
Business risk-takers
are best incentivized
by real carrots,
not just by “sticks
painted orange.”
During 1990–2000,
“Golden Carrots”
were offered for
32 new energy-saving
products in six major
categories serving
Swedish government
buyers and residential,
commercial, and
industrial customers.
Some of those
products created new
export markets.
812–817. See Box 21
on p. 200.
810a. This phrase is Steve Wiel’s when he was a Nevada Public Service Commissioner.
811. Olerup 2001.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
199
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: “Golden Carrots”
21: Golden Carrots: theme and variations
Golden Carrots (defined on p. 199) have elicited
efficient products from windows (Sweden) to
photocopiers (EPA, DOE, and International
Energy Agency—a multinational procurement
project), from lighting can fixtures and miniature
compact fluorescent lamps (Pacific Northwest
National Laboratory) to rooftop air conditioners
(DOE, PNNL, and Defense Logistics Agency).812
Golden Carrots have also been used by many
European governments, and the resulting lessons are well described.813 Sweden even undertook an exploratory battery-electric car procurement in 1994–96, including the normal
steps: an active buyers’ group, media coverage,
targeted informational materials, exhibitions,
and rewards for providers of the first product
batch. All these phases merit in-kind and financial support. Initial production support in a U.S.
context could be done by loan guarantee, purchase, or prize. Many combinations are possible: e.g., failing to win a first-past-the-post competition shouldn’t exclude other manufacturers
from loan guarantees for capacity to make
products meeting the criteria.
Golden Carrots are a special case of a broader
“Technology Procurement” approach that
needn’t rely on government procurement, but
aims additionally or instead at the general
market. The emerging international pattern is
for a neutral public body to:
812. PNNL 2003; Hollomon et al. 2002, p. 6.125; Wene & Nilsson 2003.
1. organize and aggregate selected highvolume buyers and market influencers
(such as utilities),
2. closely understand the business and technology needs of diverse and often fragmented buyers,
3. aggregate those users’ specifications into
highly desirable product attributes,
4. develop technical specifications with
buyers and makers,
5. solicit competitive bids for compliant
new products,
6. select one or (preferably) more winning
bidders,
7. help them with marketing and customer
education to maximize purchase and
ensure that diffusion takes root, and
8. keep driving further technological
improvements.
Private fleet operators, such as car-rental and
taxi companies, would be suitable participants
for such aggregations.
Another popular format is to offer a prize paid
out as the first x units are sold by cutoff time y.
In 1993, the U.S. EPA and 24 utilities, combined
into the Consortium for Energy Efficiency (CEE),
created the Super-Efficient Refrigerator
Program to improve the appliance that uses
(continued on next page)
813. Olerup 2001; Wene & Nilsson 2003;
Suvilehto & Öfverholm 1998 (whose Appendix 2 tabulates which activities have been applied to which products); IEA 2003b.
814. Frantz 1993.
815. Whirlpool apparently quietly continued making a comparable model for some time under Sears’s Kenmore label. Later, the industry agreed on
tighter federal standards from 1998, but after Whirlpool launched an effort to boost efficiency further, its domestic competitors sought and got a
three-year delay in the new standards, causing Whirlpool to quit the Association of Home Appliance Manufacturers in protest. Some would draw
the sobering lesson that while the competitors were busy using their lobbying muscle to soften the competitive edge that Whirlpool’s SERP-driven
innovation had gained, foreign competitors were quietly gaining ground on all of them—unsettlingly like the auto business.
200
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles
“Platinum Carrot” advanced-technology contest
In 1714, the British Parliament offered a huge cash prize—£10,000 to
£20,000, depending on precision—for inventing a way to measure ships’
longitude. The result prevented countless shipwrecks and revolutionized
world trade.818 In the 1780s, the French Academy offered a 100,000-franc
prize for a process to extract sodium hydroxide from sea salt, and so created the modern chemical industry. The 1927 Orteig Prize for which
Charles Lindbergh flew the Atlantic was one of about three dozen aviation prizes offered during ~1908–15. Indeed, the lure of privately funded
prizes is credited with jumpstarting the U.S. aviation industry, and the
1895 Great Chicago Car Race (based mainly on innovation, not speed)
with “giving birth to the American auto industry” 819 by engaging Charles
Duryea and his contemporaries in both cooperative and competitive
efforts at a time when American car tinkers were scattered and coherent
development was centered in Europe. This tradition continues. Round818. Sobel 1995. However, John Harrison’s precision-clockmaking effort, launched in 1730 and culminating in a half-degreeprecise clock in 1761, met with skepticism: the Longitude Board refused to believe longitude could be determined without
astronomical measures. It awarded only half the prize, then kept demanding more evidence and more clocks, until ultimately
King George III bypassed the Board and awarded the balance.
Implementation
The first firm to make
and sell a serious
number of highly
advanced vehicles
should get a
ten-figure prize,
substituting
competitive juice for
bureaucratic soup.
The lure of privately
funded prizes is
credited with jumpstarting the U.S.
aviation industry
and giving birth
to the American
auto industry.
819. Macauley 2004; NAS/NAE 1999.
Box 21: Golden Carrots:
theme and variations (continued)
about one-sixth of the electricity in American
homes. SERP offered a $30-million prize to the
manufacturer that could produce, market, and
distribute the most efficient refrigerator at the
lowest price, using no ozone-depleting chemicals and beating the 1993 federal efficiency
standard by at least 25%.814 Whirlpool beat the
13 other entrants with 30–41%, and the runnerup, Frigidaire, announced that it too would bring
its entry to market. The contest accelerated by
about six years the U.S. introduction of market816. Ledbetter et al. 1999.
817. Later, the New York Power Authority teamed with CEE and the
federal Department of Housing and Urban Development to elicit and
market an efficient apartment-sized refrigerator via all housing authorities, utilities, and weatherization providers; since 1997, more than
200,000 units with better than doubled efficiency have been installed,
kicked off by a 20,000-unit procurement by the New York City Housing
Authority (CEE 2004; DOE 2003). Such efforts need to aim high because
technology changes quickly. For example, these refrigerators were a
useful advance, but they use ~5 times as much electricity as a larger,
privately built unit in use at RMI since 1984, which in turn uses ~2–3
times as much as the 2004 state of the art.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
leading superefficient refrigerators, even
though Whirlpool didn’t collect the whole prize
(payable as the first quarter-million units were
sold) because its sales fell ~30–35% short.
The firm dropped the new model before the July
1997 sales deadline, blaming the unit’s higher
price; 815 federal evaluators cited “insufficient
and problematic marketing.”816 But units twofold
more efficient still, recently introduced by
Danish and Japanese competitors, should soon
hit the U.S. market, so Whirlpool was merely
ahead of its time and its detractors were shortsighted.817 Indeed, a general lesson of technology procurement programs is that they often
help the losing entrants too: they further refine
their products and often come back with seriously competitive products, but meanwhile
the process has put the first mover into gear.
At worst, such contests help to prepare manufacturers to compete effectively as better products, from makers at home or abroad, begin to
attack their core markets.
201
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: “Platinum Carrot”contest
the-world nonstop flight, human-powered cross-English-Channel flight,
and other extraordinary aviation achievements were induced by monetary rewards. At this writing, aerospace composites wizard Burt Rutan
and his competitors are vying for the $10-million Ansari X-Prize for the
first private space flight.820
We may have
already proposed
enough policies on
the demand side support to support
automakers’ financing leapfrogs, not just
wanting to. However,
the Big Three have
such weak balance
sheets and guarded
competitive
prospects that they
may simply lack the
creditworthiness and
liquidity to
place the needed
big bets—especially
to retool in the short
term before most of
the new market
growth has occurred.
In 1999, the National Academy of Engineering recommended that Congress encourage more extensive experiments in speeding innovation by
prizes and contests.821 As we demonstrated in the first 32 pages of this
book, the goal of bringing drastically more efficient light vehicles into
the market is a high national priority: it is important enough to warrant
a large incentive. We therefore recommend a federal “Platinum Carrot”
prize for the first U.S. automaker that can produce and sell into the market
200,000 uncompromised 60-mpg gasoline-fueled midsize SUVs (or their
attribute equivalents in other subclasses), meeting the most advanced
emission and safety standards and capable of total gasoline/ethanol fuel
flexibility. Qualifying sales could be to any U.S. customers, civil or military,
private or governmental, and may permit double-dipping with Golden
Carrot incentives, at any prices the parties agree upon.822
The prize will have to be large enough to induce several automakers to
participate, and provide for recovery of most of the full-program development costs of a next-generation vehicle, say around $1 billion. The goal is
to bring new efficiency technologies to market in uncompromised vehicles at a competitive price—hence the sales requirement. Alternatively, the
prize could be paid out per vehicle sold; effectively this would be a
$5,000/vehicle subsidy to early adopters (worth ~$10k when leveraged to
retail), but could help such vehicles work down the cost curve to become
fully cost-competitive at production scale. The prize could be winnertake-all, or could be split among the top few contestants (e.g., in a 5:3:1
ratio) according to competitive standing in sales by a certain date.
For comparison, the precompetitive 1993–2001 Partnership for a New
Generation of Vehicles spent ~$1.5 billion 823 of federal funds,824 roughly
matched by the Big Three, and its FreedomCAR/FreedomFuel successor
was announced as a $1.7-billion program ($0.72 billion new, the rest
reprogrammed) spread over five years. As Mary Tolan of Accenture
noted, “$1.7 billion over five years does not drive the private sector into
820. Additional information available at www.xprize.org.
821. NAS/NAE 1999, p. 1.
822. Entrants might choose to sell at $5,000 below cost if they were confident of making it up by winning the prize—or to discount even deeper if they felt this would win them greater value from first-mover advantage in selling enough vehicles to
catch up with their loss-leader discounts. That’d be up to them.
823. In mixed current dollars, as is the following $1.7 billion figure for 2001–06.
824. Often to much better effect than was widely reported. It developed not only useful concept cars (Fig. 10) but also significant manufacturing improvements and, most importantly, major rivalry in advanced technology vehicles, both within the
U.S. and worldwide. Reports of deepening rifts between the Big Three over reluctance to do their best work in view of their
competitors indicated PNGV’s success in this key regard.
202
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: “Platinum Carrot”contest
action” 825—but then again, a billion-dollar prize may get more respect.
Both federal automotive programs have had value. But they would be
usefully complemented, at the least, by a developer-driven, wholly competitive approach motivated by real money—a competition where all
entrants are free to take their best shot in their own way. If successful,
the prize could then be repeated with the bar set, say, ~70% higher, corresponding to a fuel-cell version. By then, still further potential, chiefly
from more advanced materials, should be evident and worth incentivizing—much as aviation prizes, in less than a century, have progressed in
distance from a 25-meter flight to a round-the-world nonstop, and in
height from a thousand feet to outer space.
Supporting automotive retooling and retraining
The five preceding initiatives aim to shift demand strongly toward
advanced technology vehicles, secure early markets for them, and reduce
or remove the market risk of their aggressive development. These initiatives may collectively be big enough to pay for automakers’ development
and conversion costs for the first one or two million vehicles within a
normal business model. We may therefore have already proposed enough
prices on the demand side to support automakers’ actually financing leapfrogs, not just wanting to.826 However, we have also argued that the Big
Three have such weak balance sheets and guarded competitive prospects
that they may simply lack the creditworthiness and liquidity to place the
needed big bets—especially to retool in the short term before most of the
new market growth has occurred. Such financial constraints on retooling
would not be a risk to incur lightly. For the competitive reasons described
on pp. 130–137, the outcomes could be worse than Chrysler’s need for a
$2.6 billion (2000 $) bailout by 1979 federal loan guarantees827 in the wake
of the 1970s oil shocks, than Ford’s close approach to a similar fate, and
than GM’s near-miss in 1992, when it was reportedly within an hour of a
fatal credit downgrade.
In the spirit of the task-force report we quoted on p. 178, we therefore
assume for present purposes that a convincing case can be made for further support. It would apply to converting production capacity, and
probably also to building new capacity, to produce State of the Art vehicles
and their distinctive key systems and components. U.S. facilities of both
automakers and their Tier One and Two suppliers would be eligible.
If and as needed, measured and temporary federal conversion support
could be structured in several ways, aimed at reducing risk and improving liquidity. We prefer a federal loan guarantee for new or retrofitted
plant and equipment installed in the U.S. (but open to all automakers),
refundable to the government if utilized and if the plant fails to operate
for an agreed period, and specifically targeted to advanced (SOA-level)
technology vehicle production, tightly defined. Associated retraining
would also qualify, again carefully defined. The budgetary cost to the
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Before advanced
vehicle markets
become large and
vibrant, hard-pressed
American automakers
may need loan
guarantees for the
liquidity to convert
their factories quickly
enough to meet
demand ahead of
foreign competitors.
825. Tolan 2003.
826. Carson 2003 and pp.
130–137 summarize the
basic issues: overcapacity,
high fixed costs (notably
health-care), pension obligations (GM’s, with more
than two pensioners per
employee, exceeds its market cap), rigidities in labor
and dealer relationships,
and a deteriorating competitive posture.
827. The loans were later
repaid, so the federal guarantee was not called, and
the government, haven
taken the usual equity
stake, was able to sell its
shares in 1983 for a $311million profit. The 1990
Federal Credit Reform Act
requires the estimated cost
of federal credit extensions
to be estimated and budgeted; the outstanding total
of federal loan guarantees
is now well into eleven
figures, though presumably
most will never be used.
Risk can be shared by
partial (rather than 100%)
guarantees and, at a higher
cost, by subordinating the
guaranteed debt.
203
Implementation
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Retooling and retraining
If and as needed,
measured and
temporary federal
conversion support
could be structured
in several ways
aimed at reducing
risk and improving
liquidity.
We prefer a federal
loan guarantee—
the support should
reward success,
not effort.
Treasury would be offset by warrants on the firm’s equity, as in the
Chrysler bailout, and risk should be shared.828 We suggest the loan-guarantee format because the borrowers, though enjoying the increased credit
and liquidity of the federal debt rating, would be on the hook for the loan
and have an incentive to invest it wisely. Direct federal financing should
be unnecessary given this alternative, but might be considered for reasons
we haven’t thought of. Tax credits may be the least attractive option, but
if granted, should be paid per vehicle produced to stringent and marketleading (not incremental) efficiency and emissions targets: the support
should reward success, not effort.829
The Pentagon can be
the biggest moneysaver from advanced
transportation technologies. Its R&D
should lead the rapid
development of the
advanced materials
industrial cluster and
other key enabling
technologies—
investing more money
in American ingenuity
to save oil, so we
can all spend much
less trying to get
and guard it.
Federal energy R&D
also needs serious
funding, restructuring,
and a focus on
best buys.
Using a federal loan guarantee to buy down an automaker’s interest rate
by about two-fifths is equivalent to cutting ~$153 off the manufacturing
cost, hence ~$300 off the retail price, of a typical State of the Art vehicle.
This is helpful, and should support the early, pre-demand-stimulus
retooling rate shown in the solid green line in Fig. 37a (p. 183). But over
the long run, it’s much less important than the demand stimuli we’ve
proposed. Nor is money the only constraint on how rapidly the industry
can change course. Retooling rate is a function of not just investment risk
perception and liquidity, but also physical, labor, and institutional constraints and (less easily analyzed but perhaps more important) cultural
constraints—how quickly people, especially in groups, can change their
attitudes and mindsets. Abundant, cheap financing cannot push these
other rate-limiting steps beyond a certain point, so it’s highly effective
up to a point, but then less so.
R&D and early military procurement
The “Military vehicles” section, on pages 84–93 above, described the compelling doctrinal, operational, and budgetary reasons for the Pentagon to
bring superefficient land, sea, and air tactical platforms rapidly to market.
The special logistical burdens of nontactical vehicles, such as the global
hauling of materials by heavy trucks and lift aircraft, make these as big a
priority as tactical platforms. Science and technology targets for rapid
insertion span across the whole gamut of military equipment. They range
from cheap but ultralight structural materials (especially advanced composites)—which can also have immediate benefits for theater force protection—to fuel-cell APUs for tanks, other armor, and heavy trucks. Current
government and private R&D, both military and civilian, generally
emphasizes propulsion systems but badly underinvests in lightweight
materials and advanced manufacturing (Fig. 19, p. 64). Most official
assessments reinforce this error by virtually ignoring platform physics,
828. Chrysler’s equity holders gained more value than the guarantee-protected debt holders (Chen, Chen, & Sears. 1986).
829. This lesson was repeatedly learned the hard way with renewable energy tax credits. Structured to reward expenditure, they attracted rent-seekers to
compete with sound firms, spoiled market acceptance with fly-by-night operators and shoddy products, and may even have retarded achievement.
Restructured for results, as for the wind production tax credit, they have proven highly successful in driving rapid technological and business learning—
except where, as for that particular credit, Congress keeps turning off the tap more or less annually, which is another effective way to kill the U.S. industry
and cede its opportunities to foreign competitors.
204
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: R&D and early military procurement
Implementation
A recent rough estimate suggested that one-sixth of all energy in the
though encouragingly, lightweightUnited States is consumed by aerodynamic drag in transportation
ing is gaining adherents even faster
systems, with potential savings around $20b/y from applying known
in DoD than in Detroit. Advanced
drag-reduction techniques. The main obstacle is severe fragmentation
materials (pp. 159–162) are the
between institutions and across disciplines.
nexus where military and civilian
needs most clearly converge and where collaboration can advance the oil
and national security needs of both communities. This makes it imperative to focus DoD’s science and technology resources far more intensively
on such key enabling technologies for saving fuel. Design competitions to
smash the comfortable boundaries of design integration and produce
breakthrough efficiency at lower cost should also prove valuable.
Basic and applied research focused across intellectual boundaries can
pay unexpectedly large dividends. For example, a recent rough estimate
suggested that one-sixth of all energy in the United States is consumed
by aerodynamic drag in transportation systems, with potential savings
around $20b/y from applying known drag-reduction techniques.830 These
estimates may be too high—we estimate that about half, not three-fifths,
of U.S. transportation energy overcomes fluid drag—but that’s still a huge
number meriting a concerted attack. The main obstacle is severe fragmentation of effort between institutions and, more seriously, across disciplines.831 That opportunity cries out for a focused national civil/military
development effort, perhaps organized by DARPA (with Defense Science
Board guidance) in concert with diverse civilian experts. Further expansion of the sometimes lively collaboration and cross-pollination between
military developers (such as DARPA and the Services’ research centers)
and other centers of excellence (National Laboratories, private firms, and
nonprofits) can also speed such technologies’ development and transfer
to wide civilian use, but will require simpler contracting mechanisms so
innovative small groups can play. Risk- and benefit-sharing can accelerate
through R&D consortia where, for example, parties keep the intellectual
property they brought, but cross-license any results of their collaboration.
Broader reforms in civilian energy R&D are also long overdue.832 Its real outlays have fallen by nearly three-fifths since 1980, bringing the U.S. into dead
last among top industrial nations in the fraction of R&D devoted to energy.833
The private sector, far from picking up the slack, is generally slashing even
more muscle from its R&D budgets; few Wall Street analysts seem to realize
that most industries are eating their seed corn and are gravely underinvesting in R&D. Sound civilian energy R&D priorities are also unlikely to
emerge from the Department of Energy so long as it remains driven by traditional constituencies and distracted by its mainly nuclear-weapons and
waste-cleanup responsibilities; it should be restructured as purely civilian
agency with a purely energy-focused mission.834 Its R&D on manifestly
uncompetitive energy technologies (such as fission and fusion) should be
halted, greatly increasing the funds available for options that actually show
and have proven their promise. DOE’s priorities should be realigned to
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
830. Wood 2004. Further
scrutiny may refine the striking initial estimate that 60%
of U.S. transportation energy
is dissipated by fluid drag.
The cost estimates appear
dubious in detail but in the
right general direction.
831. Including (Wood 2004)
“aerodynamics, transportation,
hydrodynamics, wind engineering, environmental
sciences, chemistry, medical
engineering, combustion,
manufacturing, etc.”—and
we’d add, materials and,
most importantly, biomimetics.
832. PCAST 1997.
833. Margolis & Kammen 1999.
834. Nuclear weapons programs, for example, could
go to DoD and nuclear waste
management and cleanup
to EPA.
205
Implementation
Federal regulators
are fixing some old
auto-safety problems
but threatening to
make new ones
that are as dangerous
for automakers
as for their customers.
New materials and
designs can save
both lives and oil—
if allowed to.
Our feebate-centered
policy recommendations should gradually
render moot the
details of CAFE rules,
because virtually all
vehicles will beat
CAFE handily.
These reforms could
be gracefully phased
in about as quickly as
feebates will eliminate
their bite by profitably
shifting customer
demand toward
efficiency.
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: R&D and procurement
emphasize energy efficiency more than supply, to match the nation’s energy
needs (chiefly for decentralized heat and mobility fuels, not centralized electric power generation), and to strengthen integrative whole-systems insights
and design processes.
Automotive efficiency and safety regulation:
first, do no harm
Our feebate-centered policy recommendations, with their targeted supplements (pp. 191–209), should gradually render moot the details of CAFE
rules, because virtually all vehicles will beat CAFE handily. However,
we meanwhile endorse the Academy’s call to abolish those rules’ archaic
car/light-truck distinction and dual-fuel loophole. And to the extent feebates don’t meet and beat CAFE standards, and further CAFE reforms
are required, here’s our menu of CAFE reforms. We would treat all light
vehicles identically in safety regulations (not relax or except the rules for
light trucks); abolish the 8,500-lb limit that now lets “heavy light” passenger vehicles like the Hummer evade many existing rules; expose all light
and “heavy light” vehicles to the gas-guzzler tax (which could, however,
be repealed as feebates took effect); 835 extend both efficiency and safety
regulation to all Class 3–4 (“light medium”) vehicles; repeal the scandalously quadrupled business tax break for vehicles over 6,000 lb; 836 and
purge any other incentives for automakers to make vehicles heavier and
less efficient. These reforms could be gracefully phased in about as quickly as feebates will eliminate their bite by profitably shifting customer
demand toward efficiency.
More broadly, NHTSA’s 2004 CAFE rulemaking must not inhibit or penalize progress toward light-but-safe vehicles (pp. 57–60) by shifting, as now
proposed, from fleet-average to weight-based class average efficiency
standards that would reward heavier vehicles with looser efficiency
requirements and penalize downweighting (except for the heaviest vehicles) with stricter standards. Incentivizing weight would damage national
energy security, public safety, and automakers’ export prospects, and
would so clearly fly in the face of the scientific and engineering evidence
before NHTSA837 that it would invite challenge as arbitrary and capricious. Box 22 suggests ways to reinforce what NHTSA is doing well and
correct what it’s doing badly.
835. Substituting feebates would incidentally eliminate the gas-guzzler tax’s discrimination against cars and in favor of light trucks—much as Germany’s 16
states, for example, have recently equalized vehicle taxes between these two classes.
836. This figure is further distorted by the practice of counting SUVs’ weight including their rated payload, which is excluded for normal purposes of regulating cars. Under the Jobs and Growth Act 2003, “the entire cost of all but one large SUV—the Hummer H1—can be deducted” in the first year by any business buying one for ≥50% business use, up to $100,000; for the H1, the total deduction is $106,000 out of the $110,000 price (Taxpayers for Common Sense
2003). Many tax advisors now steer light-vehicle buyers to a list (e.g., Bankrate.com 2003) to ensure that their proposed purchase is heavy enough to qualify
for the maximum “business” tax break. If any business tax break for light vehicles is desired, it should be at least neutral as to weight, and ideally calibrated
to efficiency per unit of interior volume or footprint area (not weight). Meanwhile, things have come to such a pass that on many residential streets in
California and elsewhere, thanks to old 6,000-lb weight limits, most of today’s larger SUVs are already technically illegal (Bowers 2004).
837. Including Lovins 2004a.
838–839. See Box 21.
206
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal policy recommendations for light vehicles: Efficiency and safety regulation
Implementation
NHTSA should mine its data for the engineering reasons why, at any given weight,
some vehicles are several times as crashworthy as others: these design differences hold the key to huge,
low- or no-cost safety gains without changing cars’ manufacturing or materials at all.
22: Realigning auto-safety policy with modern engineering
NHTSA assumes a weight/size relationship as
an inflexible fact, when actually it’s a major
technical policy variable that policy should be
designed to influence strongly; yet neither
NHTSA nor the NRC report on which it relied
has considered the new materials that can radically decouple size and strength from mass. The
agency’s weight-based approach rests on logically and technically invalid extrapolation from a
dubious historic correlation (note 317, p. 60) to
the future causality of risk in fleets with different designs and, probably very different materials. This led NHTSA to adopt the false assumption that heavier means safer. If the aggressivity
and compatibility issues of recent years are any
indication, that way lies an even more absurd
mass arms race in which you drive a Hummer,
she drives an 18-wheeler, and he drives a locomotive. Such a future would ensure, not only a
greater toll on the roads, but also the isolation
and ultimately the failure of American automakers’ offerings within more discriminating global
markets whose safety rules are based on a rigorous systems engineering approach to crash
safety. As we noted on p. 60, the dean of industry safety research, retired GM safety-analysis
leader Dr. Leonard Evans, though long a vociferous critic of CAFE standards, agrees that lighter
but stronger materials can save fuel, emissions,
and lives simultaneously.
838. NHTSA should also intensify its historically weak efforts to distinguish mass from size in its historic crash-safety data; conduct a
convincing public analysis to check if its historic weight/risk correlations are artifactual (note 317, p. 60); and most importantly, launch
an open public process, in collaboration with European and
Japanese lightweight-vehicle safety experts, on how the light-butsafe safety philosophy emerging abroad (note 301, p. 58) can best
be refined, allowed, and encouraged in a U.S. context.
NHTSA should stimulate innovation in making
all (and particularly large) vehicles lightweight
and crashworthy and compatible with all other
vehicles. NHTSA should mine its data for the
engineering reasons why, at any given weight,
some vehicles are several times as crashworthy
as others: these design differences hold the key
to huge, low- or no-cost safety gains without
changing cars’ manufacturing or materials at
all.838 Sound science, the rapid emergence of
ultralight-vehicle structural and manufacturing
options, and common sense all point to the need
for a federal safety policy that respects markets, encourages innovation, promotes public
health, and advances national competitiveness.
Such a policy should:
• be performance-based, not prescriptive;
technology-neutral; as far as possible technology-forcing; and supportive, not destructive, of national efficiency goals;
• decouple fuel-economy choices from vehiclesize-class choices (if desired) by encouraging
size/mass decoupling, via normalizing fuel
intensity to size (e.g., gal/mi per interior ft 3)—
not to mass;
• be at least neutral as to vehicle mass, carefully avoiding any incentives for a further
spiral of the “mass arms race”; 839 and
• favor the downward rather than upward harmonization of mass within the fleet (if one
does wish to influence mass).
(continued on next page)
839. M. White 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
207
Implementation
Crafting coherent supportive policies
Modest federal
policy tweaks
can similarly
accelerate other
kinds of oil-saving
technologies and
practices throughout
the economy.
Other federal policy recommendations
New technologies,
bioenergy sources,
trade rules, and
policies can systematically revitalize
rural and small-town
America while
protecting soil,
community,
and climate.
• Direct DoD to conduct a “fly-off” competition as it does for military
hardware, aimed at building 5–10 commercial-scale demonstration
plants within five years and testing them to determine which processes
work best. This would be funded with $1 billion to be flexibly spent at
DoD’s discretion. Such a competition should greatly accelerate the early
plant phase of development that the private sector would otherwise
undertake more timidly, and would be a cheap way to gain the needed
technological learning on a timescale that better fits national urgency.
840. EFC 2003, “App. A:
Working Group Reports:
Report of the Bioenergy
and Agriculture Working
Group.”
841. For comparison, the
2002 Clean Coal Technology
Initiative allocated $2 billion
over 10 years to R&D for
coal—a massive, profitable, and centuries-old
industry. The 2002 Farm Bill
allocated $75 million for
biofuel R&D over 6 years.
Supporting investment in domestic energy supply infrastructure
There is a rapidly emerging consensus that accelerating cellulosic biomass
conversion and other biofuel production, if done right, is an important
part of any oil-displacement strategy. The Energy Future Coalition’s
Bioenergy and Agriculture Working Group, representing diverse stakeholders, recommends four sensible initiatives:840
• Increase federal bioenergy R&D funding from $0.15 to $0.5b/y; 841
allocate it more on technical and less on political considerations
(Congressional “earmarks” have recently overallocated increased
hydrogen funding, but with scant regard to technical merit); and
emphasize applied fundamentals.
• Redirect agricultural export subsidies—to be purged under the 1 August
2004 World Trade Organization agreement—to developing biofuel
markets in ways that conserve soil and that encourage, measure, and
reward carbon fixation in soil. The National Research Council should
assess the impacts of shifting crop subsidies to soil conservation, energy crops, and the bioenergy industry. (Some private assessments have
lately suggested that the phaseout of subsidies would suffice to put
the bioenergy industry on a sound footing as a durable source of
income to revitalize farm, ranch, and forest economies and cultures.)
22: Realigning auto-safety policy with
modern engineering (continued)
Meanwhile, NHTSA’s commendable efforts of
the past few years to reduce the aggressivity
and incompatibility of high, heavy vehicles (such
as large SUVs and pickups), and to improve their
stability, should be strongly encouraged and
accelerated. Innovations like Honda’s Advanced
Compatibility Engineering Body Structure—
already on the market in Japan and about to
enter the U.S. market—prove the effectiveness
208
and economy of such redesign. During the transitional period, while very heavy “light” vehicles
continue to be offered, and while light (though
not necessarily smaller) vehicles increase their
market share, such design reforms will be especially important to public safety. Of course, driver behavior, such as intoxication and seat-belt
usage, is enormously more important to safety
than vehicle design, and should continue to be
seriously addressed by NHTSA and lawenforcement agencies.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Other federal policy recommendations: Domestic energy supply infrastructure
Implementation
• Use government policy to increase the use of bioproducts and assess
their benefits, for example:
In Sweden, the government is suggesting
a requirement for half
of the biggest filling
stations to offer at
least one renewable
fuel by 2008, and will
soon require public
procurement to buy
efficient fuel-flexible
cars and fill them
with renewable fuels.
• Adopt a national renewable-fuels standard842 and an electricity
renewable portfolio standard.
• Expand existing tax incentives for renewable energy production
(e.g., windpower’s Production Tax Credit) to include environmentally acceptable but wasted biomass.
• Give other fuels with equal or better lifecycle environmental performance the same tax treatment afforded to [corn-derived] ethanol
under current law (i.e., create a level playing field between all biofuels).843
• Develop USDA labeling for bio-based content, and encourage government procurement of such products.
• Increase EPA’s efforts to promote alternative transportation fuels
in ozone nonattainment areas.
The only Working Group recommendation we disagree with is that
automakers should continue to receive CAFE credits for “flex-fuel” vehicles. Although our feebate proposal should rapidly render CAFE irrelevant, we think any requirements for the capability to use alternative fuels
should be separate from those for fuel economy. In addition:
• As we noted earlier (p. 107), within about a decade, the fuel flexibility
of at least E85 and preferably Brazilian-style “total flex” vehicles should
be the norm for new U.S. light vehicles; otherwise cellulosic ethanol
supply, blended or neat, could outrun the fleet’s ability to use it.
• The official definition of “biodiesel” should be broadened to include
all neat or blendable diesel fuels made from wastes; the current congressional definition, written to advance certain crop producers’ interests, uses a narrow chemical definition and feedstock requirement
(favoring virgin and edible feedstocks over used and inedible ones),
and even excludes bio-oils blendable with petroleum diesel fuel.
The best waste streams to begin with are those already being collected,
especially if they now incur a disposal cost.
In Sweden, whose farming and forestry sectors provide a strong platform
for biofuels, the government is suggesting a requirement for half of the
biggest filling stations to offer at least one renewable fuel by 2008, and will
soon require public procurement to buy efficient fuel-flexible cars and fill
them with renewable fuels. Many of the larger Swedish cities already offer
free parking for hybrids and flexible-fuel vehicles, exempt them from the
forthcoming (2005) Stockholm congestion charge, and reduce by 20% the
usage tax on company cars if they’re environmentally preferred. Sweden
and other European countries also have well-advanced plans and propos-
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
842. The Senate has passed
a 5-billion-gal/y-by-2012
ethanol standard, meant
mainly to boost corn
ethanol production from its
current 3 billion gal/y. We
think 5 billion gal/y is likely
to be exceeded anyhow, but
would like to encourage the
rapid development of cellulosic ethanol. Three good
methods would be to set
the bar much higher, e.g.,
25 billion gal/y by 2015 (10%
of U.S. motor fuel), perhaps
twice that by around 2020;
to establish a credit trading
system to elicit least-cost
solutions; and to count multiple credit, phased down
from an initial value of say
3:1, for each gallon of cellulosic ethanol in order to
reward early adopters, then
taper off that incentive as
the technology matures.
More broadly, since innovation by definition will produce surprises, we urge
its broad encouragement by
encouraging all kinds of
conversion of waste
streams into mobility fuels,
emphasizing those compatible with existing infrastructure. Otherwise, biofuels
policy will continue to be
balkanized by the rentseeking behavior of specific
technology advocates,
channel owners, and feedstock producers.
843. We expect these subsidies would trend downwards, and would not need
to exceed subsidies to gasoline if gasoline’s price properly reflected its social cost.
209
Implementation
Crafting coherent supportive policies: Other federal policy recommendations: Domestic energy supply infrastructure
als for filling stations that provide, as some already do, a diverse and flexible mixture of fuels: a normal gasohol (E5–E10) or similar gasoline-ethanol
blend that any unconverted car can use (such as Brazilian-style E22), E85,
hydrous ethanol (with due care in cold climates), various biodiesel blends,
and hydrogen. The menu would gradually evolve away from oil, toward
renewables, and toward a higher hydrogen/carbon ratio. Rural filling
stations where ethanol is available, but natural gas is not, could advantageously choose deliveries of just ethanol and gasoline (ultimately just
ethanol) and, as demand justifies, could reform some of the ethanol onsite
to hydrogen for fuel-cell vehicles.844 Perhaps direct-ethanol fuel cells
will make the reforming altogether unnecessary.845
America’s soil
conservation
programs and biofuel
and renewable
energy development
programs do comply
with WTO rules.
844. Reforming ethanol
requires a higher temperature than natural gas but
lower than gasoline, and
not only allows but welcomes water content,
which can even be much
higher than would be permissible as a direct vehicular fuel.
845. An autothermal ethanol
reformer has been demonstrated (Deluga et al. 2004).
Our other previous biofuel recommendations (pp. 107–110) include:
changing the Conservation Reserve Program rules to allow perennial notill cropping of soil-holding energy crops, especially in polyculture; promoting low- and no-input production of such feedstocks; and establishing
recommended-practices standards for domestic-biofuel production, and
labeling and tracking for imported-biofuel production, so that bioenergy
helps to protect and enhance tilth, not degrade it. We also urge great care
in developing, licensing, and applying genetically modified organisms for
biofuel production: some being created, especially those that can digest
both C5 and C6 sugars, could do immense mischief if they escaped, propagated, and perhaps swapped their genes into other organisms. Current
U.S. regulatory structures are not up to this task: national regulation of
GMO crops was indeed structured so that nobody is responsible for ensuring public health and safety—especially the manufacturers that wrote
the rules.846
Encouragingly, biofuels can nicely reconcile legally acceptable ways to
strengthen rural economies with the 1 August 2004 World Trade
Organization’s agreement to phase out export and crop subsidies in
developed countries (notably the U.S. and EU). (These subsidies were
deemed to be an impermissible bias against developing-country farmers.)
Support for rural economies can now be done not for its own sake but as
fair compensation for providing valuable public and private nonagricultural services: from renewable energy supply to soil protection to carbon
sequestration. A recent legal analysis found that America’s soil conservation programs and biofuel and renewable energy development programs
do comply with WTO rules, because these programs: have clear environmental and conservation objectives, don’t distort global trade through
direct price supports, and meet certain program-specific criteria such as
minimum periods of land set-asides. However, to insulate these programs
from WTO attack, the analysis recommends that Congress legislatively
confirm the programs’ clear environmental and conservation purposes
and document their environmental benefits.847
846. Lovins 1999.
847. Dana 2004.
210
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Other federal policy recommendations
Heavy-vehicle policy
We already proposed proven regulatory changes that could greatly
increase the purely technological oil savings available from heavy trucks
(p. 74, note 367; p. 75, note 369), including:
• Raise federal Gross Vehicle Weight Rating (GWVR) to the European
norm of 110,000 lb, trailer length from 53 to 59 ft, and trailer height
from 13.5 to 14 ft (some states have already done these or more).
Implementation
Simple reforms,
many already adopted
by states or foreign
countries, can greatly
improve Class 8
trucks’ efficiency and
profitability while
safeguarding public
health and safety.
• Allow double- and triple-trailer combinations if accompanied by disc
brakes that increase braking power per pound of GWVR so as to
improve safety.
• Remove obstacles to expanded stacking-train rail and rail-to-truck
transloading, so railways are better used for long hauls via expanded
truck-rail-truck intermodality.
• Reduce heavy-truck speed limits to 60 mph.
• Remove any regulatory obstacles to consolidating loads with large
carriers.
Properly done, these should sustain or improve safety and not increase
(possibly decrease) road wear. There might be a modest need to accelerate
repair of the most deficient bridges, but this should be done anyway for
public safety.
Objective national research should also seek to resolve over the next few
years whether advanced filtration, computer controls, and other methods
can make light-vehicle diesel engines acceptable in California and other
areas with strict air quality standards (especially for ultrafine particulates). German automakers seem to think so; as mentioned earlier, they’ve
persuaded Chancellor Schröder to ask the EU to accelerate by three years
the new Euro-5 standards, which from 2010 could cut fine particulates
by up to 99%. EPA, which has been tightening its standards as medical
research reveals more unpleasant surprises, is skeptical and may require
considerably lower emissions; so may California regulators, who are
allowed to set their own stricter standard. In any event, resolving this
issue is a vital prerequisite to rethinking emissions from future dieselpowered heavy vehicles, and hence for the U.S. role for biodiesel (though
not for ethanol). If diesel technology turned out to be unacceptable for
long-run public health, such alternatives as fuel cells would need to be
accelerated. Meanwhile, welcome recent initiatives are starting to clean
up off-road, railway, and marine diesels through better technology,
lower-sulfur fuel, and reduced idling (substituting APUs or dockside
hookups)—Canadian locomotives turn out to idle 54–83% of the time.848
Welcome recent
initiatives are starting
to clean up off-road,
railway, and marine
diesels through
better technology,
lower-sulfur fuel,
and reduced idling.
848. WBCSD 2004, p. 92.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
211
Implementation
Crafting coherent supportive policies: Other federal policy recommendations
Regulatory reform
in fuel, gate, and slot
pricing can give
airlines a fair shot at
success and improve
the quality of travel.
Aircraft policy
Charging drivers
what driving really
costs, making
markets in avoided
travel, and designing
communities around
people, not cars,
can make access
and mobility cheaper,
fairer, and more
pleasant.
The best way not to
need to travel is to be
already where you
want to be, so you
needn’t go somewhere else.
Real-time,
congestion-based
road pricing
is a good replacement for lost gasoline
tax revenues.
Aviation fuel’s treaty-bound freedom from taxation may be coming to an
end, none too soon. We suggest that this distortion (competing mobility
fuels are taxed, often heavily) be corrected by environmental user fees, on
the emerging European model. Zürich and some other European airports
have also pioneered variable but revenue-neutral landing fees calibrated to
the noise, emissions, or other nuisances of the various types of aircraft, so
as to signal correctly their relative public value. Nearly half of landings at
Zürich now pay less than before, while the most polluting planes pay 35%
more in a steep graduated fee structure meant to accelerate cleanup.
Besides the loan-guarantee/scrappage program suggested on p. 191, we
suggest policy encouraging a level playing field for hub-and-spokes vs.
point-to-point business models. In practice, this would mean allocating
gates and slots through the market rather than letting them continue to be
hoarded by “fortress hub” monopolists.849 If that means the monopolists
can’t compete with point-to-point carriers, that should be their problem,
not their customers’.
Other transportation policy
A fully rounded transportation policy to foster affordable mobility with
less or no oil is not all about vehicles and fuels. It must also deal with
dwindling gasoline tax revenue (as vehicles become more efficient, and
perhaps switch to other fuels that may not be taxed the same as gasoline),
improved intermodal transportation, and smart growth.
The core of any sensible transportation policy is to allow and promote
fair competition, at honest prices, between all ways of getting around or
of not needing to. The best way “not to need to travel,” i.e. to generate
“negatrips,” is to be already where you want to be, so you needn’t go
somewhere else. That could mean more sensible land-use, plus virtual
mobility that moves only the electrons while leaving the heavy nuclei
behind.850 Another key element of transportation policy is to ensure equitable access for all Americans, in ways that improve customer choice,
the quality of our communities, and the air we all breathe.
Shifting taxation from fuel to roads and driving
More efficient vehicles will reduce fuel purchases and hence motor-fuel
tax revenues, even if all forms of fuel, including biofuels and hydrogen,
are taxed comparably to gasoline (which would be counterproductive and
hard to justify from economic principles, since gasoline taxes are supposed to reflect externalities, not a mere desire for government revenue).
849. The airport nearest RMI’s headquarters is dominated by such a firm, with the result that—in between the sporadic periods it had competitors over the
past two decades—the airline has sometimes charged more to fly 25 minutes to its fortress hub in Denver than to fly from there to Europe.
850. Many transportation policy achievements and proposals are surveyed in Hawken, Lovins, & Lovins 1999, pp. 40–47.
212
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Federal: Transportation: Shifting taxation from fuel to roads and driving
Offsetting this tax loss will require other ways to fund building and
repairing roads, bridges, and other highway infrastructure.851 To be sensible, forward-looking, and economically efficient, user fees charged on
roads or driving should be real-time and congestion-based: if government
is charging such fees, it might as well do it in a way that reflects the true
social cost of taking a trip right now. Signaling that cost will reduce the
need to build and maintain infrastructure in the first place, as well as save
time, accidents, and pollution. Signaling that cost also allows fair competition by other modes of physical mobility and, indeed, by other means
of access, such as substituting virtual for physical mobility, or co-location
to design out the need for travel.
There is extensive and encouraging experience, particularly in Europe, with
real-time and congestion-based tolling on highways for both passenger and
freight vehicles.852 Another handy tool is cordon pricing in cities, similar to
the policy recently introduced in London, which cut traffic by 16% within
months, and to those used in several Norwegian cities and in Singapore.853
U.S. experiments with electronic tolling and other forms of automatic road
charges 854 are encouraging, and the interoperable telemetric devices emerging throughout the Northeast could be adapted to real-time congestionbased pricing. The same is true of parking fees 855 and bridge and ferry tolls:
if they’re charged electronically, the cost of adding time and congestion
information is usually small. Some privately built highways even charge
a toll based on real-time congestion, but guarantee its refund if a “service
quality guarantee” (so many minutes to a given destination) isn’t met; then
drivers can decide if reduced time-of-travel-uncertainty is worth the price.
A realistic, cost-based approach to driving would for the first time send
efficient market signals to drivers, reducing sprawl, travel, congestion,
pollution, and oil use. It would ensure that drivers not only get what they
pay for, but also pay for what they get. And it would improve equity
between drivers and the one-third of Americans—a growing class as our
demographic grays—who are too old, young, poor, or infirm to drive.
Sooner or later, this “immobilized class” will start insisting that drivers
pay their own way without burdening the taxes of those who can’t drive
and thus end up paying twice for mobility—once for the socialized costs
of others’ car-based mobility, and again for their own alternatives.
Implementation
User fees charged
on roads or driving
should be real-time
and congestion-based.
A realistic, cost-based
approach to driving
would ensure that
drivers not only get
what they pay for,
but also pay for what
they get. And it would
improve equity
between drivers
and the one-third of
Americans who are
too old, young, poor,
or infirm to drive.
851. To offset gasoline-tax revenues lost to greater fuel economy (p. 41), we had to assume some unspecified mix of user fees totaling ~24¢/gal-equivalent to
sustain 2004 revenues and 62¢/gal to achieve revenues scaled to EIA’s 2025 forecast. Some ways of doing this are more efficient, effective, and palatable
than others, but the choice should be local and isn’t ours to make.
852. Perkins 2002; Viegas 2002; Perett 2003.
853. IEA (2001) estimates that cordon pricing in all major metropolitan areas could cut 2010 light-vehicle consumption by ~3–6% (IEA 2001, p. 15). Acceptance
by the public is typically raised by linking toll-ring revenue to improvements in public transportation and transport infrastructure.
854. IEA (2001, p. 110) and California’s SR-91 experience suggest that allowing toll-paying drivers with low occupancy to use HOV lanes, so they become
“HOT” (High Occupancy/Toll) lanes, increases those lanes’ utilization and acceptance, and that their usage is increased mainly by a diverse set of drivers
suffering occasional urgency rather than by wealthy drivers using them routinely.
855. Overprovision of seemingly free parking is the biggest single cause of U.S. urban congestion. Parking cash-out—so driving competes more fairly with
other modes of mobility and access—is a useful solution (IEA 2001). If combined with a U.S. national parking tax of $1/h (up to $3/d), parking cash-out is estimated to be able to cut light-vehicle travel, fuel use, and CO2 by 4–7% by 2010 (IEA 2001, p. 16).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
213
Implementation
Crafting coherent supportive policies: Other federal policy recommendations: Other transportation policy
A whole-systems
approach to transportation should
focus on intermodal
competition, integration, and innovationrisk management.
Integrating transportation systems
The Houston-sized
city of Curitiba has
the second-highest
car ownership in
Brazil but the lowest
car drivership, the
cleanest urban air,
and the highest
quality of life.
Avoiding unwanted
mobility is the most
powerful long-run
focus for transportation policy.
856. CyberTran
(www.cybertran.com),
for example, being an ultralight rail system manyfold
cheaper, more compact,
and more flexible than conventional light rail, holds
promise for retrofit within
as well as between cities.
Unfortunately it falls
through the cracks between
conventional categories, so
it has long been unable to
attract the program-defined
government R&D funding it
merits. The technology was
originally developed at the
Idaho National Engineering
Laboratory as an alternative
to the hour-long bus commute to the remote site for
experimental nuclear reactors, but that didn’t create a
niche for it within DOT.
214
A fair transportation policy would also include improved intermodal
and mass transit options such as Bus Rapid Transit and improved light
rail. (Western Europeans use their well-integrated and convenient public
transport for a tenth of their urban trips; Canadians, 7%; Americans,
2%. The Houston-sized city of Curitiba has the second-highest car ownership in Brazil but the lowest car drivership, the cleanest urban air, and
the highest quality of life.) It’s unrealistic to expect most American cities
to approach downtown Tokyo’s 92%-rail commuting, but uncivilized to
have a rail system that many cities in the developing world would scorn.
With new technologies ranging from CyberTran® to hybrid-electric bicycles, American industry could indeed zoom ahead of foreign competitors—if the U.S. government had the flexibility to recognize and support
such innovations’ emergence,856 e.g., by insuring against first-adopter
technical risk, much as EPA used to do for innovative wastewater systems. (If the novel wastewater system didn’t work, despite the expectations of EPA’s technical experts, EPA would pay for replacing it with a
conventional one, so the early adopter could seek the benefits without the
risk.) Similar risk management could break the logjam on commercializing technologies like CyberTran.
Is this trip necessary (and desired)?
Sound policy would also include encouraging smart growth designed to
reduce sprawl and the need for driving. Most importantly, it would stop
subsidizing and mandating sprawl, so we would have much less of it.
Personal mobility in America is often as undesired as it’s excessive.
It comes increasingly from the deliberate segregation and dispersion of
where we live, work, shop, and play, and its effect is to fragment our
time, erode our communities, weaken our families, and raise our taxes to
pay costs incurred by developers but socialized to everyone. Americans
travel more miles than all other industrial nations combined, and U.S.
light vehicles emit as much greenhouse gases as all of Japan. This is neither necessary nor economic, and it may not be making us happier.
Indeed, designing cities around cars, not people, makes them unlivable.
And it is hardly a worthy model for the rest of the world, which as a
whole is only half as good at birth control for cars as for people.
Finally, a good transportation policy would tackle problems in freight
transportation, creating a more efficient intermodal system that would
shift freight transport from trucks to rail and/or ships as much as possible and worthwhile. A comprehensive, economically efficient transportation policy for both goods and people is vital, but its design is outside the
scope of this study, and our savings calculations assume none of it.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: Other federal policy recommendations
Non-transportation federal policy
The main oil-saving national policies we suggest outside the sphere of
transportation are few and simple. For example, the same discount-rate
arbitrage that we suggest for vehicle buyers would also make sense in
buildings and industry, so anyone considering buying energy efficiency
will take as long a view of its benefits as society does. The many ways to do
this include: proposed laws (such as S.2311 and H.R. 4206 in 2004) offering
comprehensive tax incentives for substantial long-lived efficiency improvements in equipment and buildings; 857 broader and faster building and
equipment efficiency standards; further improvements in and enforcement
of procurement policies; fundamental improvements in how buildings and
equipment are designed; 858 feebates for buildings (as a few jurisdictions now
do for water and sewer hookups); educating real-estate-market actors about
the valuable labor-productivity, retail-sales, and financial-return benefits of
efficient buildings; 859 and applying widely an innovation now used to make
the capital market for energy savings efficient by providing a standard way
to measure, aggregate, and securitize them just like home mortgages.860
Federal energy policy should take a coherent approach to mobility and
access, land-use, and safety. A more difficult but equally important shift
would be to ensure that our energy prices tell the truth—shorn of subsidies
and reflecting external (larcenous) costs now imposed on people at other
times and places than the energy point of sale. We hope we live to see such
honest prices. Another critical policy objective would be to encourage and
permit all ways to produce and save energy to compete fairly, no matter
which kind they are, what technology they use, how big they are, or who
owns them. That fundamental reform would be revolutionary in a system
long attuned to constituents’ needs rather than the broad public interest.
With or without these basic reforms, a more subtle, seldom noticed, and
perhaps even more important federal opportunity offers arguably the
greatest single leverage point in national energy policy: correcting a glaringly perverse incentive in the retail price formation of regulated gas and
electric distribution utilities in 48 states, as noted in the next section. This
principle could be federally articulated and encouraged,861 even though its
implementation is largely a matter for the states, (including Territories
and Tribes), to whose responsibilities and opportunities we turn next.
Implementation
Proven methods of
pricing, finance, and
design can help use
energy in ways that
save money. An even
more revolutionary
notion: let all ways
to produce or save
energy compete fully
and fairly.
The same discountrate arbitrage that
we suggest for
vehicle buyers would
also make sense in
buildings and industry, e.g., via feebates
for buildings and by
applying widely an
innovation now used
to make the capital
market for energy
savings efficient:
by providing a
standard way to
measure, aggregate,
and securitize them
just like home
mortgages.
857. This is not just for
saving electricity: PG&E
estimates that by 2001,
standards and utility incentives had cut California’s
direct use of natural gas
(not for power generation)
by more than one-fifth.
858. RMI’s 10XE (“Factor Ten Engineering”) project aims to change fundamentally how engineering is done and taught, so as to achieve radical—order-ofmagnitude—energy and resource savings, usually at lower capital cost and with improved performance. Hawken, Lovins, & Lovins 1999 give examples.
859. Note 541 and Innovest 2002, which found one-third better stock-market and financial performance by leaders in energy management within the commercial property sector. Similar studies for the retail merchandising and retail food sectors are found at Energy Star, undated.
860. Such as the International Performance Measurement and Verification Protocol, www.ipmvp.org, now used nearly worldwide to finance energy and
water savings, both new and retrofit, in the private, public, and nonprofit sectors.
861. Before the current expansionist tendencies of the Federal Energy Regulatory Commission’s jurisdiction—defined by the 1932 Federal Power Act as
relating solely to interstate power transactions—there was ample and often encouraging precedent for federal laws that required each state at least to consider certain utility regulatory reforms. Unfortunately, in recent years some state and federal courts have flatly ignored key provisions of such landmark federal laws as the National Utility Regulatory Policies Act (1978), on the strange principle that since some states have restructured their electricity markets,
PURPA must be a dead letter nationwide, even though Congress forgot to repeal it.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
215
Implementation
In such areas
as utility and
insurance regulation,
car- and fuel-tax
collection, and
land-use, states and
localities are emerging as leading
innovators.
If the federal government won’t follow,
it should get out of
the way and not
reward the opposite
of what we want
(paying oil companies
to find oil, the
Pentagon to protect
it, and a business
owner to buy a
Hummer to waste it).
States have a large
and effective
menu of ways to lead
in transportation
reforms.
Crafting coherent supportive policies
States: incubators and accelerators
If the federal government doesn’t lead, it should follow by letting the
states try experiments, then perhaps standardizing what would work best
nationwide. And if the federal government won’t follow, it should get out
of the way: not preempt state experiments that create public goods (from
Maryland feebates to California CO 2 reductions), not favor the worst buys
first, and not reward the opposite of what we want (e.g., paying oil companies to find oil, the Pentagon to protect it, and a business owner to buy
a Hummer to waste it).
It is constitutionally axiomatic, but often forgotten, that under federalism,
powers not specifically reserved to the national government are reserved
to the states. The states have their own challenges, but on the whole, their
vast diversity and greater grassroots vitality tend to make their governments more creative, dynamic, and accountable than the often-gridlocked
federal government. Moreover, states can interact more directly with communities—an even more accountable and often effective level of action
than state government. Thus a galaxy of extraordinarily effective energysaving actions were conceived and executed at a city, county, town, village, and neighborhood level all across America in the 1970s and early
1980s, often encouraged by state energy offices funded by the U.S.
Department of Energy.862 Those successes, now forgotten by all but oldtimers,863 offer an encouraging template for reforging the links between
government, civil society, and the ultimate engines of action—the social
atoms of the firm, household, and individual.864
Transportation
The Northeast Advanced Vehicle Consortium in New England,865 and
the network of sophisticated California agencies with their Northeastern
counterparts who prefer California’s to federal air standards, illustrate
strong state leadership in transportation. They provide a framework for
implementing feebates at a regional or state level if federal action falters.
As we noted earlier (p. 189), federal preemption could be avoided by
careful legislative drafting, even if the federal government refused to
waive the CAFE authority whose legally mandated bar-raising it’s been
refusing to implement. States are also free to implement scrappage programs, as some already have.866
862. At that time, DOE had a lively local-programs office led by Tina Hobson (now Senior Fellow at Renew America). Federal funds were later slashed by
more than 80%, but Congress still provides ~$0.3b/y. The National Association of State Energy Officials (www.naseo.org), with a vast reservoir of field experience, illustrates the extent of this underinvestment by noting that Alabama’s commercial/industrial efficiency program generated more than $750 for each
dollar spent. Only a few states, such as California and North Carolina, have durably institutionalized their energy efforts, and many are at annual risk of disappearing into budget-deficit crevasses.
863. Outstanding examples are reviewed by Alec Jenkins in Lovins & Lovins 1982, Ch. 17, pp. 293–334.
864. Political theory has a lot to say about the relationship between the individual and the nation-state, but is relatively weak on the structure and function of
communities. RMI’s Economic Renewal Program has long found communities a peerless arena for social innovation.
865. Additional information available at www.navc.org.
216
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies: States: incubators and accelerators: Transportation
States are the almost exclusive arbiters of the ground-rules that govern
zoning, land-use, road pricing, and other basic underpinnings of travel
demand. Those policies can either incur or avoid the “enormous costs
of sprawl,” which a Bank of America study found “has shifted from an
engine of California’s growth to a force that now threatens to inhibit growth
and degrade the quality of our life.” 867 States can also encourage local
use of sprawl-reducing mortgages, such as Fannie Mae’s Location Efficient
Mortgage and Smart Commute products,868 to achieve major increases in
disposable income and reductions in traffic.869 In addition, states can:
• Adopt Pay-at-the-Pump collision liability insurance (Box 23, p. 218)—
an excellent candidate for national standardization once state and
regional models have been refined. It should save ~$11 billion a year.
• Supplement national or state feebates with direct auto-dealer incentives. The sales associate who pockets a $100 special state bonus for
selling a superefficient car, on top of normal commission, will push
such sales more enthusiastically.
• Pay carrying charges for dealers to stock superefficient but not ordinary cars. This would incentivize automakers to fill the inventory
pipeline with diverse advanced-technology models, especially in areas
served by smaller dealers. Otherwise, initially thin sales might reduce
advanced vehicles’ public exposure by making them a special-order
item that sells poorly because there are none on the lot.
• Realign light vehicles’ registration fees not so much with market value
as with lifecycle value—market value plus present-valued lifetime fuel
consumption—so the public-goods value of efficient vehicles is re-signaled annually.870 (Even better would be to include clear externalities.)
The current ad valorem system encourages drivers to hang onto old,
inefficient cars rather than buy new, efficient ones.
866. However, some further state actions remain ensnarled. Many states have long wanted to offer hybrid cars free access
to HOV lanes, but most have concluded this is illegal until EPA corrects a regulatory mistake that makes such discretionary
access—meant to save oil—depend not on the vehicles’ fuel economy but solely on whether they burn an approved alternative fuel.
867. Bank of America et al. 1996
868. Modeled on Fannie Mae’s Energy Efficient and Home Performance Power products, which finance energy-saving
home improvements (new or retrofit) and apply the proceeds to mortgage qualification ratios, these products increase borrowing power for homebuyers whose location near workplaces or transit can reduce their commuting costs. These innovations are all part of Fannie Mae’s ten-year, $2-trillion American Dream Commitment ® to bring homeownership to an additional 18 million Americans in targeted groups.
869. Studies in three cities found that, compared with sprawl, higher urban density cuts driving by up to two-fifths, proximity to transit by one-fifth (Holtzclaw et al. 2002).
Implementation
A galaxy of extraordinarily effective
energy-saving
actions at a city,
county, town, village,
and neighborhood
level all across
America in the 1970s
and early 1980s offer
an encouraging
template for reforging
the links between
government, civil
society, and the
ultimate engines
of action—the social
atoms of the firm,
household, and
individual.
The sales associate
who pockets a $100
special state bonus
for selling a superefficient car, on top of
normal commission,
will push such sales
more enthusiastically.
Realign light vehicles’
registration fees not
so much with market
value as with lifecycle value—market
value plus presentvalued lifetime fuel
consumption—so the
public-goods value of
efficient vehicles is
re-signaled annually.
870. Implementing the OECD “Polluter Pays Principle” (which the U.S. officially accepts in principle), most European car
ownership taxes depend on some combination of engine type and displacement, fuel type, vehicle age, and gross vehicle
weight (in Sweden, Norway, and the Netherlands). British and French ownership taxes depend on CO2 emissions.
Denmark’s move from a weight- to an inefficiency-based tax in 1997 has cut fuel intensity by 0.5 km/L (1.2 mpg) for gasoline
vehicles and by 2.3 km/L (5.4 mpg) for diesel vehicles. Emissions may also be considered, as in Tokyo and some other
Japanese prefectures, to correct the current perverse incentives that clean new vehicles are taxed at far higher rates than
dirty old ones (Hirota & Minato 2002).
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
217
Implementation
Crafting coherent supportive policies: States: incubators and accelerators: Transportation
23: Pay-at-the-Pump car insurance
Since both insurance and state fuel taxation
are state issues, a key opportunity for state
leadership to save light-vehicle fuel and reduce
the cost of driving, especially for low-income
citizens, is Pay-at-the-Pump (PATP) automotive
liability insurance.871 Current third-party auto
collision insurance costs the same no matter
how few or many miles are driven.872 Many drivers buy no insurance, usually because they
can’t afford it, and hope they won’t get caught
(most don’t, though there may be consequences).873 Those who drive less or play by the
rules thus subsidize both high-mileage and uninsured drivers. Moreover, since low-income
people drive only about half as many miles as
the well-to-do, they pay about twice as much
per mile, thus cross-subsidizing the rich—and
paying not in affordable slices but in unaffordably large chunks.874 PATP was suggested as far
back as 1925, and Senator Moynihan tried to
include it in his 1967 reforms.875 With the socialized costs of uninsured motorists (and resulting
litigation) rising, this idea’s time has finally
come, in a hybrid form that melds the best of
the PATP and traditional insurance payment
systems, and should overcome the objections
sometimes raised to previous versions.
Specifically, we propose that basic third-party
property-damage and bodily-injury insurance be
bought at the fuel pump via the existing state
fuel-tax system and repaid to each state’s insurance issuers in proportion to their current-year
market share. Other insurance and extra coverage would be paid to one’s chosen company
just as now, trued up for any risk premia or for
competitive differences between insurers. This
is simply a smarter way to pay about one-third
of your insurance bill, and reduces everyone’s
bills because there are no longer any uninsured
218
motorists (you can drive without insurance, but
not without fuel). Uninsured motors would be
automatically assigned to a carrier pro rata on
their market share. Insurance companies would
gain more customers with no marketing effort;
PATP should sustain or improve their profitability.
PATP also reminds drivers every time they refill
the tank that part of the variable cost of driving
is exposure to collision risk, and that, like fuel
use, this cost can be reduced by driving less.
To this extent, the variabilized price signal would
be more efficient than the present flat-rate lumpsum.876 It might seem at first that drivers of
efficient cars get cross-subsidized by drivers of
(continued on next page)
871. Khazzoom 2000. A well-known popular treatment (Tobias 1993)
entangled the concept with no-fault and partly public insurance—
among the main reasons the insurance industry successfully opposed
it in every state where a campaign was run at the time.
872. Although accident rates don’t vary linearly with miles driven, they
do depend more on congestion, which is related to collective driving:
by definition, traffic density equals vehicle-miles divided by lane-miles
of capacity. Obviously, too, driving more miles exposes you to more
potential accidents.
873. Collision liability insurance or its equivalent is mandatory throughout the United States. The Insurance Information Institute (www.iii.org/
individuals/auto/a/canidrive) states that NH, TN, and WI require only
financial responsibility, FL requires only property-damage liability coverage, and the other 46 states and District of Columbia require both
bodily-injury and property-damage liability coverage.
874. The last survey data (1993), from EIA 1997’s Household Vehicles
Energy Consumption 1994 (this survey was discontinued in 1994 but
may perhaps be revived), showed 49% fewer miles/vehicle-y at household incomes <$15,000 than for those ≥$50,000 (Khazzoom 2000a, p. 26).
Interestingly, the respective average vehicle efficiencies were 19.8
and 20.1 mpg. Not surprisingly, a 1993 survey of 799 low- and moderate-income Californians found 89% in favor of PATP—96% among the
majority without insurance (Khazzoom 2000a, p. 25)—and the concept
was strongly supported by low-income advocacy groups (Khazzoom
2000a, p. 24, note 38).
875. Khazzoom 2000a.
876. Unlike “Pay as you Drive” insurance—another way to variabilize
the cost—there’s no need to check how many miles each car travels
(via periodic odometer checks or real-time GPS or other telemetry),
and all drivers are covered to the extent they drive, rather than leaving
some with the option of driving uninsured. The PATP portion of insurance premia would be automatically adjusted up or down with fleet
fuel economy. It also marginally incentivizes fuel economy.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Crafting coherent supportive policies: States: incubators and accelerators: Transportation
• Encourage localities to grant parking preferences to efficient cars,
as some American and European cities already do.
• Help their highway departments buy rubberized asphalt early and
often to capture the states’ dominant share of the $8b/y of agency
savings on paving costs. Some of these savings may then be recycled
into accelerated bridge upgrades and repairs (p. 211).
Electricity and natural-gas pricing
Saving natural gas, largely by saving electricity, is a key element of displacing oil, so we must pay attention to all three forms of energy, not just
the direct use of oil. States have original and unique jurisdiction relating
to all three, such as intrastate transportation and the retail pricing of electricity and natural gas. Almost every state has a utility commission that
regulates jurisdictional private (and sometimes, in part, public) utilities;
many still regulate distribution utilities’ formation of retail prices. Due to
historical accidents, all states except Oregon and California currently do
this in a way that rewards utilities for selling more energy and penalizes
them for cutting customers’ bills. The National Association of Regulatory
Utility Commissioners, with rare unanimity, resolved in 1989 that this
perverse incentive should be fixed, and nine or so states did so. By 2000,
all nine except Oregon reversed this reform (often inadvertently), as they
became distracted by restructuring or succumbed to short-term political
pressures to freeze electric rates (prices)—a conceptual blunder, because
customers pay not rates but bills (price multiplied by consumption).
Box 23: Pay-at-the-Pump car insurance (continued)
inefficient cars, but in fact, charging per gallon,
not per mile, also has a rational basis: the vehicular efficiency being rewarded will increasingly
come from lightweighting, which reduces
aggressivity toward other vehicles and toward
pedestrians. The combination of PATP with our
earlier proposals for low-income access to efficient vehicles is particularly attractive for both
equity and economic efficiency.
Implementing PATP today would add a typical
insurance charge of ~$0.45/gal at the pump,
but reduce insurance bills by more than that:
the average annual fixed cost per car would fall
by ~$250 (2003 $), and the total cost of driving
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
States hold the most
potent U.S. energypolicy lever:
aligning retail utilities’
financial interests
with their customers’.
All states except
Oregon and California
currently reward
utilities for selling
more energy
and penalizes them
for cutting
customers’ bills.
would fall.877 The California Energy Commission
estimates a net effect of reducing light-vehicle
fuel consumption by 4% by 2020, saving California drivers $1.3 billion in direct non-environmental costs (which would scale to ~$11 billion
nationwide) in present value.878 We already
counted the appropriate net effects of PATP in
calculating net driving rebound (p. 41, above).
PATP would best be implemented nationally,
but even if it were done in just a state or region,
past experience with different taxes suggests
only modest and declining “retail leakage” from
some drivers’ tanking up out-of-state to avoid
the premium.
877. Khazzoom 2000a; Ashuckian et al. 2003, pp. 3-4 – 3-9.
878. Ashuckian et al. 2003.
219
Implementation
Crafting coherent supportive policies: States: incubators and accelerators: Electricity and natural-gas pricing
The solution is arrestingly simple: use
a balancing account
to decouple utilities’
profits from their
sales volumes, then
let utilities keep as
extra profit a small
part of what they
save their customers.
This is the most
important single way
to make natural gas
and electricity cheap
and abundant again.
Then California, chastened by the costly consequences of nearly destroying its world-leading efficiency programs, restored its successful old solution, and other states are now starting to follow suit.
States can
diversify their
energy supplies to
include renewables
in a way that
decreases their
citizens’ costs
and risks.
The solution is arrestingly simple: 879 use a balancing account to decouple
utilities’ profits from their sales volumes, so they’re no longer rewarded
for selling more energy nor penalized for selling less, and then let utilities
keep as extra profit a small part of what they save their customers, thus
aligning both parties’ interests. This is the most important single way to
make natural gas and electricity cheap and abundant again. It can be supplemented by state financing, typically by revolving funds to invest in
efficiency and renewables: the largest of these, San Francisco’s $100-million bond issue, is expected to pay for itself with a profit to the taxpayers.
Honolulu’s $7.85-million revolving fund for City buildings’ energy
improvements is expected to earn $2 million net.
Renewable energy
Recent federal gridlock has hatched more state energy leadership and
spread it into regional compacts with greater scale and leverage. In mid2004, even as the federal government pushed fossil-fuel development
across the Western states, the nine states of the Western Governors’
Association, led by Governors Schwarzenegger (R-CA) and Richardson
(D-NM), unanimously agreed to develop 30 GW of clean (generally renewable) electric capacity by 2015, raise energy efficiency 15% by 2020, and
seek to become “the Saudi Arabia of wind and solar energy.” 880 Thirteen
states have officially adopted mandatory “renewable portfolio standards”
for their electricity or energy supply portfolios; Governor Pataki (R-NY)
has set a 25% goal for the next decade. This is not just prudent diversification but also sound financial economics, because renewables like solar and
wind energy, especially when diversified and aggregated over a substantial area, can be bought on fixed-price contracts with no material price risk
to the developer (nobody hikes the price of sun and wind, so once installed,
it’s a constant-cost resource within the modest statistical fluctuations of
weather and microclimate). An optimized portfolio should therefore
include renewables in the energy supply portfolio even if they appear to
cost more per kilowatt-hour, for the same reason that an optimized investment portfolio should include riskless Treasury debt even if it yields less.881
And in fact, some renewables, such as well-sited windpower, are now
clearly competitive on price alone, thanks partly to state and federal
public-goods R&D that’s helped to launch new industries.
879. Some of the best retired state utility regulators, via the nonprofit Regulatory Assistance Project, advise their peers how
to do it both in the U.S. and abroad: www.raponline.org.
880. Cart 2004. Additional information available at www.westgov.org.
881. This argument, elaborated chiefly by Dr. Shimon Awerbuch, is summarized and documented in Lovins et al. 2002, pp.163–167.
220
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies
Military policy:
fuel efficiency for mission effectiveness
As the rest of the federal government leads, follows, or gets out of the
way, there’s one part of the government that is trained, prepared, and
obliged to lead: the Department of Defense. We showed on pp. 84–93 why
DoD absolutely requires superefficient land, sea, and air platforms to fulfill its national-security mission. To be sure, as a byproduct of platform
efficiency, DoD will also greatly enhance its warfighting capability and
trim probably tens of billions of dollars per year off its overstressed budgets; but the most important result will be the Pentagon’s ability to deploy
and sustain agile, effective forces. The more the military can be relieved of
the duty to protect oil, the safer will be our troops, our nation, and the
world. And the military’s spearheading of lightweighting and other efficiency technologies will greatly hasten the day it will be relieved of this
unnecessary mission.
Right now, the Pentagon’s requirements-writing, design, and procurement
of military platforms place a modest rhetorical priority but a low actual
priority on fuel efficiency. The Defense Science Board (DSB) task force’s
report of 31 January 2001 set out the needed reforms.882 It was briefed to
the full DSB in May 2001, released mid-August 2001, and concurred with
by the Joint Chiefs of Staff 24 August 2001.883 Eighteen days after that
came 9/11, which diverted attention from any further action. Decisive
adoption is now overdue. Fortunately, there are growing signs, chiefly
within the Services, of interest in shifting in-house and contractor design
professionals toward integrative, whole-system thinking that can “tunnel
through the cost barrier” to radical energy efficiency at lower capital
cost.884 The main missing ingredient is turning efficiency aspirations into
actual requirements and acquisitions. Leadership to do so must come
from the Secretary of Defense.
This need is driven partly by a growing realization among the senior uniformed and civilian leadership that cost is a strategy—that organizations
and processes must deliver requisite capabilities dramatically quicker and
cheaper, not just in first cost but in the total long-term cost of using what
we buy. A strategic approach to cost must therefore emphasize strategies
not just for sharply reducing acquisition and operating costs (often by
changing metrics, e.g. properly counting logistics), but also by suppressing
the monetary cost of war, countering adversaries’ strategies for imposing
cost, and reducing the likelihood and intensity of war itself (p. 261). As the
fuel-efficiency strand of that strategy emerges, DoD could also take practical steps with shared-savings pilot tests under the proposed National
Defense Energy Savings Bill.885
Implementation
Light, agile forces
can’t do their job if
they drag a heavy
logistical tail. But
innovative techniques
promise to transform
this vulnerability into
major warfighting
gains, tail-to-tooth
realignments, fiscal
savings, and civilian
spinoffs.
The more the military
can be relieved of the
duty to protect oil,
the safer will be our
troops, our nation,
and the world.
And the military’s
spearheading of lightweighting and other
efficiency technologies will greatly
hasten the day it will
be relieved of this
unnecessary mission.
882. DSB 2001, p. 87,
note 426.
883. Fry 2001. The only nonconcurrence was over
whether “fuel efficiency”
should be a mandatory Key
Performance Parameter
expressed in operational
requirements documents
and whose nonattainment
is cause for program termination; the Joint Staff
believed “fuel efficiency
should not adversely
impact meeting an operational requirement.” We
don’t interpret this subordination of fuel efficiency to
operational requirements
as a denigration of fuel efficiency’s great importance,
and believe it can be
required in other effective
ways than as a KPP.
884. Lovins 2003a; Hawken, Lovins, & Lovins 1999, especially Ch. 6.
885. S.2318 cosponsored by Senators Collins (R-ME), Bayh (D-IN), Roberts (R-KS), and Reed (D-RI); see the Alliance to Save Energy 2004.
The bill died when CBO “scored” its obligations without its contractually offsetting savings. CBO and OMB should fix this unwarranted practice.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
221
Implementation
Crafting coherent supportive policies
Civil preparedness: evolving toward resilience
In evolving
away from today’s
overcentralized and
highly vulnerable
domestic energy
system, efficient use
buys the most bounce
per buck, and more
diverse, dispersed,
renewable supplies
can make major
supply failures
impossible by design.
Meanwhile, homeland security is at risk as intelligent adversaries threaten to exploit the profound and longstanding vulnerabilities of America’s
overcentralized, brittle energy systems (p. 12).886 Our centralized systems—
the regional power grids, oil and gas pipeline networks, and the like—
could be brought down by single acts of sabotage; indeed, large pieces of
the power grid keep failing all by themselves. Building more giant power
lines and power stations will make this problem worse, because at its
root, the problem is one of system architecture, not of inadequate capacity: the bigger they come, the harder they fall. However, the demandresponse strategy we proposed on pp. 111–122, initially as a quick way to
save trillions of cubic feet of natural gas per year at negative net cost, also
happens to make the electricity system more resilient in much the same
way that many distributed, instantly usable fire extinguishers help to
protect a flammable building. Dispatching even a small amount of load
management into a transmission system that’s about to tip into cascading
blackout can instantly correct the supply/demand imbalance before it
can propagate. Whether one is designing precautions against terrorists or
tree-limbs, saboteurs or squirrels, instant local response from the demand
side (and from intelligently self-dispatching distributed generators), in an
omnidirectional grid, is a key missing element of energy security.
Another basic finding of our 1981 Pentagon analysis of domestic energy
vulnerability and resilience (p. 12) is that the most “bounce per buck”
comes from using energy more efficiently. For example, tripled-efficiency
light vehicles with the same number of gallons in their fuel tanks can
run three times as far before refueling. If their fuel supply is disrupted,
that leaves three times as long to mend it or to improvise new supplies.
Alternative supplies can also support three times as much mobility,
greatly increasing flexibility.
In the tightly coupled oil supply chain, where crude-oil stockpiles like the
Strategic Petroleum Reserve are many steps and miles from users’ needs
for refined products, and where major stocks of products are equally
vulnerable to attack, the most important way to buy resilience is end-use
efficiency. Thus efficient vehicles (and other oil uses) act as a fine-grained,
highly distributed Strategic Petroleum Reserve—already delivered to
customers, presenting no high-value targets, invulnerable to cascading
system failures (such as vulnerable pipeline networks), and profitable
to boot. That’s an element of energy security and national resilience that
no people in a dangerous world can afford to be without.
The most important way to buy resilience
is end-use efficiency.
886. Lovins & Lovins 1982;
Lovins et al. 2002; Lovins, Datta, & Swisher 2003.
222
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies
Implementation
Civil society: the sum of all choices
For the past 95 pages we’ve analyzed American business opportunities
and policy needs. But the bedrock supporting those major actors is the
awesome power of more than 280 million diverse citizens who, given
good information, have shown for two and a quarter centuries an uncanny knack for ultimately making farsighted choices. For all the ills and
errors of our society, the power of its democratic roots and its rich networks of interlocking communities is immense, and growing ever more
so in the Internet age. Despite initial concerns that the information revolution would only empower tyrants and spread darkness with the speed of
light, the record so far suggests that more people have already been liberated by the microchip than ever were by the sword. And as communities
of home and work, interest and faith, begin to coalesce and interlock,
linked electronically in new networks that erase the isolation of distance
and status, citizens’ power becomes a rising tide that can drive great
social movements. The more diverse the elements of that power, the
greater the power of the whole.
Just as informed citizens drive politics, so informed customers drive business. For all the concern about manipulation by concentrated media and
skillful advertising in both the civil and commercial spheres, any business
executive knows that customers are ultimately in control. The franchise,
the license to do business, depends critically on public approbation. One
of the biggest challenges of leading a large organization is that anyone in
it, in a thoughtless moment, can do something stupid that makes the public unwilling to buy that company’s products. Such a reputational loss can
in the end prove fatal. The converse opportunity is to be the kind of company people are eager to deal with. And any “green CEO” knows that the
biggest win from earning a sterling reputation, and from removing any
contradiction employees might have felt between what they do at work
and what they want for their kids, is an unassailable lead in recruitment,
retention, and motivation of their people.
By mindful market
choices, collaboration with innovative
firms, and disapprobation of laggards,
civil society can
drive business and
government innovation as powerfully as
it did in the California
electricity crisis,
when informed and
mobilized customers
undid 5–10 years’
previous demand
growth in just
six months.
In such a world, where reputation is such a precious but fragile asset, civil
society has an unprecedented power to reward the leaders and punish the
laggards in making business serve not just narrow economic ends but
also broad societal ones. Nongovernmental organizations, already a sizeable chunk of the economy and perhaps the leading source of social innovation, amplify this bottom-up power of engaged citizens to help steer
business in creative new directions.
Because we lead better than we follow—more a bunch of lone wolves than a flock of sheep—we are greater together
than we could ever be apart, able to lead and innovate from the bottom up as well as any people on earth.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
223
Implementation
Crafting coherent supportive policies: Civil society: the sum of all choices
Some citizens organize to work with companies; others, to combat them.
But whether civil society and corporations coevolve in a predator-prey
relationship, warily circling each other, or overtly cooperate in innovation, they are the two most important partners in today’s tripolar world
of business, civil society, and government. Increasingly, civil society and
private enterprise are teaming up to do together what government should
do but can’t or won’t. In new patterns, we find new ways to join together
and learn together.
In this kaleidoscope of ad-hoc collaborations and shifting alliances, leadership from both sides and all levels is vital, and goes far beyond mere management. (As Admiral Grace Hopper said, “Manage things; lead people.”)
But the most important leadership still comes from the people, from the
urban neighborhood and the village square, the schoolroom and the factory, the church and the Little League field. It comes from the conveniencestore counter and the truck stop, from the seat of a tractor and the back of
a horse. We Americans are an adventurous, independent, and ornery lot.
Because we lead better than we follow—more a bunch of lone wolves
than a flock of sheep—we are greater together than we could ever be
apart, able to lead and innovate from the bottom up as well as any people
on earth. Even top-down leadership can elicit that bottom-up power if
done with proper respect. As Lao Tze said,887
Leaders are best when people scarcely know they exist,
not so good when people obey and acclaim them,
worst when people despise them.
Fail to honor people, they fail to honor you.
But of good leaders who talk little,
when their work is done, their task fulfilled,
the people will all say: “We did this ourselves!”
887. In the classic Tao Tê Ching, #17.
224
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Crafting coherent supportive policies
Beyond gridlock: changing politics
In our society, citizens are not mere passive consumers but the wellspring
of emerging values and opinions that drive business and political evolution. The silent but inexorable motion of belief systems and alliances, like
the shifting of tectonic plates, can cause sudden realignments. In crafting
an approach with broad appeal across the political spectrum, we are mindful of the potential for new and unexpected coalitions to form:
• Advocates for elders, the handicapped, the sick, and the impoverished
can join environmentalists in favoring equitable access to mobility for
all, especially non-drivers.
Implementation
Realigning policy
with public needs
can forge broad and
powerful new coalitions around clear,
common-sense
objectives and the
values shared across
a land of diverse
patriots.
• Quality-of-life and community advocates can join anti-tax conservatives in insisting that developers pay their own way, rather than burdening the rest of the community with higher taxes for the road,
school, public-safety, and other costs their projects impose.
• Conversely, canny real-estate developers who know the unique profitability of neotraditional and “new urbanist” design can join with
both environmentalists and family-focused conservatives to create
great places to raise families. These are designs that focus on pedestrian neighborhoods, not traffic engineering, so that as Alan Durning
puts it, cars can become a useful accessory of life rather than its central
organizing principle.
• The progressive environmental/labor axis can join with both hardnosed industrialists and faith communities committed to stewardship
in new ways of doing business that, as Ray C. Anderson says, “do
well—very well—by doing good.”888
• Rural and farm advocates can find their most deeply felt needs—
cultural as much as economic—converging with those of four possibly
unfamiliar allies: industrialists building new bioenergy enterprises,
environmentalists protecting land, water, and climate, military leaders
seeking new ways to strengthen the roots of national security, and
advocates for the world’s impoverished farmers, seeking fair trade
undistorted by old-style American crop subsidies.
Such lists can be almost indefinitely expanded. We offer these examples
of seemingly unlikely bedfellows not as amusing artifacts but as a stimulus to thought about the new, trans-ideological politics of a path beyond
oil based on common sense, shared interests, and good business.
888. Hawken, Lovins, & Lovins 1999.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
225
Implementation
Crafting coherent supportive policies: Beyond gridlock: changing politics
Too long has our nation been riven by energy-driven conflicts between
regions and cultures—Midwestern coal-miners and -burners vs. friends of
Northeastern lakes and trees, oil- and gas-exporting vs. -consuming states,
automaking powerhouses vs. oil-dependent and climate-sensitive states,
the urban core vs. the suburbs, old-timers vs. immigrants. Our divisions
spread deeper, too, when diverse cultural roots have been nurtured in
different soils: for example, between those not versed in country things
and those who can say, in farmer-poet Wendell Berry’s words, that “What
I stand for is what I stand on.” There are many causes of these divisions.
But in innumerable ways, both obvious and subtle, our oil habit has been
an important part of what has divided us.
Our immense national diversity—an undying source of strength and
delight—will never wholly transcend such differences of interest, emphasis, and agenda. But as many of the underlying causes of those tensions
are revealed as artifacts of an obsolete and uneconomic energy strategy,
those causes can increasingly be sublimated into a healthy source of
cohesion. We can then begin to collaborate on the great national project
of the 21st Century: building together in America a secure, convivial,
and prosperous life after oil.
226
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Option 4. Substituting hydrogen
Beyond mobilization
to a basic shift in primary energy supply
On pp. 123–125 we summarized the technical means to cut 2025 oil
demand by more than half. In the Implementation section (pp. 127–226)
we explained in business and policy terms how to achieve these goals,
thereby breaking OPEC’s pricing power and indeed making oil relevant
only to countries that choose not to adopt similarly profitable approaches
themselves For the U.S., this takes three integrated steps: saving oil at less
than half its price, biosubstitution at up to the oil price, and substituting
saved natural gas at levels that make sense regardless of the oil (and gas)
price. The remaining oil demand, 13.5 Mbbl/d, could then be provided
by the 7.8 Mbbl/d of forecast domestic oil output plus 5 Mbbl/d of
“balance.” That “balance” could come from any combination of six other
resources: waiting for the additional 7 Mbbl/d of cheap efficiency already
underway to be fully implemented, importing North American biofuels
or oil, eliciting more efficiency or biosubstitutes by starting to count oil’s
unmonetized costs, or substituting more of the 8 TCF/y of leftover saved
natural gas via hydrogen or compressed-natural-gas vehicles. Three other
points bear emphasis:
• Of the 2025 oil demand, 4.4 Mbbl/d will fuel cars and light trucks.
After 2025, State of the Art vehicles will continue to be adopted until
they start to saturate the market. They’ll be two-thirds of the lightvehicle stock by 2035, when all transportation will use only 12.9
Mbbl/d (extrapolating EIA’s demand forecast beyond its end in 2025).
The supply/demand balance will thus be even better in 2035 than
in 2025—total demand down to 10.6 Mbbl/d, the remainder, after
domestic supply substitutions, down to 3.3 Mbbl/d. And it will still
keep improving, with a rich menu of options to choose from.
• If we hadn’t saved half the oil by end-use efficiency, then trying to do
the same thing with supply-side substitution alone, notably biofuels,
would ultimately have run into supply constraints and very high
prices (Fig. 30, p. 103) as well as competition with food crops and
constraints in water availability and land-use. These issues, often cited
by critics of biofuels, are artifacts of very high demand for mobility
fuels. This is unnecessary if we use a least-cost combination of efficient
use and alternative supply.
Option 4.
Superefficient light
vehicles create a robust
business case for hydrogen-fuel-cell propulsion.
Deployed in concert
with fuel cells in buildings, this permits a
rapid, practical,
profitable, and
self-financing
hydrogen transition,
a durable all-domestic
energy supply strategy,
and many other
important benefits.
Oil-related carbon
emissions would fall,
then cease.
The need for natural gas
and financial
capital would be modest
and may shrink.
Efficient oil use,
biosubstitutes, and the
obviously profitable
substitutions of saved
natural gas for oil can
halve oil use by 2035
despite rapid economic
growth, but the
trajectory of oil use
can be aimed
irreversibly toward zero
by an advantageous
(though not essential)
further option—
making the leftover
saved gas, or other
abundant energy
such as windpower,
into hydrogen.
• Examining the structure of our tools for modeling the effects of policies,
we estimate that a modest fraction of the Mobilization scenario’s oil
savings (~35% of light-vehicle savings and an unknown fraction for
other end-uses) might occur anyway through spontaneous market
uptake of incremental technologies. However, we definitely don’t recWinning the Oil Endgame: Innovation for Profits, Jobs, and Security
227
Implementation
Three other points
(continued):
The leftover saved
gas could be used
even more effectively
than simply burning it
as a direct fuel.
If it were instead
converted to hydrogen, it could make
U.S. mobility
completely oil-free.
889. Lovins 2003b. DOE
states the factor is 2.5
(Garman 2003). (Burning H2
in an internal-combustion
engine, instead of using it
to run a fuel cell, reduces
its efficiency advantage
over hydrocarbons to
~1.3–1.5 and compromises
its economic case.) In stationary applications, onsite
H2 cogeneration or trigeneration typically offers at
least twice the efficiency of
a central fossil-fueled
power station plus an
onsite combustion heater.
Option 4. Substituting hydrogen: Beyond mobilization to a basic shift in primary energy supply
ommend such inaction, because it would forego most of the benefits.
The total of benefits available from fully implemented Mobilization
(State of the Art technologies accelerated by Coherent Engagement policies), for the projected 2025 level of services, includes $133 billion a
year in avoided oil purchases, or a net benefit of $70 billion a year. It
also provides all the resulting side-benefits, such as reduced exposure
to oil-price volatility, lower global oil tensions, reduced emissions,
deferred depletion, etc., plus the benefits of industrial and rural revitalization. Even under the same policies, incremental Mobilization technologies (in the Let’s Get Started scenario) would achieve less than half
of these benefits at more than half of their cost. Such a failure of nerve
would leave far too much money on the table.
We know that optimal results depend both on advanced oil-displacing
technologies and on policies to support their rapid adoption. But a question
remains: is this as good as we can do, or is another step beyond today’s
State of the Art technologies already coming into view as technical progress,
accelerated by the policies we’ve suggested, continues its virtuous spiral?
A further important potential step—the hydrogen transition—would be
optional but advantageous. We noted on p. 124 that the 8.2 TCF/y of leftover saved gas could be used even more effectively than simply burning it
as a direct fuel. If it were instead converted to hydrogen, which can be
used 2–3 times as efficiently,889 then it could meet the transportation needs
within both the 5 Mbbl/d “balance” requirement and the 7.8 Mbbl/d of
domestic oil output if desired, making U.S. mobility completely oil-free.890
Since hydrogen can be produced from any other form of energy, including
renewables, hydrogen also holds the key to the long-term elimination of
oil use for transportation (even including airplanes) and all other uses.
We thus stand poised for the final checkmate move in the Oil Endgame, the
move that can deliver total energy independence—the hydrogen economy.
To illustrate its power, we show an altered version of Fig. 33 from p. 123.
This time, in Fig. 38, we convert the 8 TCF/y of leftover saved gas into
hydrogen. This graph demonstrates that even if hydrogen realizes only
10% stock capture by 2030, it already would make a significant difference
in oil consumption because it eliminates oil use entirely. There are so many
attractive ways to make hydrogen that this illustrative 10% can be
increased as much as desired.
890. This also provides
an “insurance policy” in
case the oil savings from
efficiency and biofuels so
depress world oil prices
that the costlier parts of the
7.8 Mbbl/d of U.S. oil output
become uncompetitive
and get shut in or left undeveloped.
Once hydrogen is introduced,
the complete elimination of oil use is inevitable;
the only question is how fast
this would happen.
228
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Option 4. Substituting hydrogen: Beyond mobilization to a basic shift in primary energy supply
Figure 38: The path beyond oil becomes even more flexible if the hydrogen option is added
As usual, the difference between petroleum use and petroleum imports is EIA’s forecast of domestic petroleum production, which by
law excludes all areas now off-limits. That oil output isn’t adjusted here for slower depletion, which in practice would make domestic oil
resources last longer or permit oil from particularly high-cost or sensitive areas to be left untouched. The gray line illustrates the first
10% stock capture of all end-uses by hydrogen over the decade starting in 2025, with far more still to come.
petroleum product equivalent
(Mbbl/d)
35
30
government projection
(extrapolated after 2025)
25
end-use efficiency
@ $12/bbl
petroleum use
20
plus supply substitution
@ <$26/bbl
15
petroleum imports
10
5
2030
2020
2010
2000
1990
1980
1970
1960
1950
0
plus optional hydrogen from
leftover saved natural gas
and/or renewables
(illustrating 10% substitution;
100%+ is feasible)
year
Source: EIA 2003c; EIA 2004; preceding RMI analysis.
Thus, once hydrogen is introduced, the complete elimination of oil use is
inevitable; the only question is how fast this would happen. For primary
fuel substitutions, substituting one primary fuel for another, such as the
successive shifts from wood to coal to oil to gas, has historically required
approximately 50 years for each shift, although they overlapped. Other
complex technology substitutions such as railways or electrification that
required technical interdependence, significant infrastructure, and large
scale have also required 40–50 years. There are some compelling reasons
to believe that hydrogen could displace other fuels more quickly. The integration of fuel cells 891 into vehicles and buildings will be made far simpler
by the technological and efficiency improvements already in place if the
State of the Art technology portfolio is implemented in a timely fashion.
Coherent Engagement policies will already have removed barriers, and
aligned incentives for introducing distributed hydrogen technologies and
for accelerating the development and adoption of successive generations
of hydrogen fuel cells. As we’ll discuss below, there will be an abundance
of diverse primary energy sources for the hydrogen transition. These factors
suggest to the authors that hydrogen could largely or wholly displace oil
25–30 years after it is introduced. Embarking on the hydrogen transition,
whatever its exact timing, sets the trajectory of oil use unmistakably and
permanently downward.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
891. The fuel cell, invented
in 1839, is a modular electrochemical device. Many
high-school chemistry students do an experiment in
“electrolysis,” using an
electric current to split
water into hydrogen and
oxygen. A fuel cell does the
same thing backwards: it
chemically recombines
hydrogen with oxygen (typically from the air) on a catalytic membrane, with no
combustion, to produce an
electric current, pure hot
water, and nothing else.
There are several kinds of
fuel cells. Those expected
to be first applied in vehicles typically use direct
hydrogen produced separately by “reforming” a
hydrocarbon or carbohydrate with steam. Some
higher-temperature types of
fuel cells can use hydrocarbon fuels directly.
229
Implementation
Option 4. Substituting hydrogen: Beyond mobilization to a basic shift in primary energy supply
Hydrogen could
largely or wholly
displace oil
25–30 years after
it is introduced.
Fig. 38’s flexible and profitable recipe for a vibrant U.S. economy without
oil is illustrative. Such a mapping of how complementary elements could
unfold helps to provide insight into technological and policy opportunities, not to predict or prescribe a precise future. Interactions between
energy supply, demand, and price are traditionally simulated by learned
specialists using large econometric or general-equilibrium computer models.
While the feedbacks reflected in dynamic models are often real, not mere
artifacts or circular logic, they generally capture too much or too little
(or both) of the energy/economic system’s immense complexity. For this
reason, and because we pledged a transparent system-level analysis using
downloadable spreadsheets so readers could scrutinize and change our
assumptions (p. 33), we unapologetically offer simpler but not simplistic
scenarios. They are neither forecasts nor fantasies. They fall, as Professor
Paul Steinhart used to say, “between the unavoidable and the miraculous.”
They seek more to enable the future than to predict it (pp. xxiv, 34).
Despite recent
skepticism by
the underinformed,
industries and
governments are
pressing ahead with
billions of dollars’
worth of R&D that is
making the hydrogen
transition practical
and profitable.
Hydrogen: practical after all
In that exploratory and indicative spirit, we sketch here the main elements
of a hydrogen transition that is optional—we can displace oil profitably
even without it—but would be advantageous in profits, emissions,
fuel flexibility, and security. It would also reduce calls on natural-gas
resources, whether by stretching them longer or by avoiding those of
highest private and public cost. Ultimately, hydrogen would make possible a completely renewable solution to mobility fuels, so they could not
be cut off, would never run out, and wouldn’t harm the earth’s climate.
Done well, these fuels could also cost less per mile and have steady,
predictable prices.
Along the way, we briefly address some of the salient misunderstandings
that have led many otherwise well-informed commentators to criticize the
hydrogen economy as infeasible, uneconomic, dangerous, or polluting.
These misconceptions are dealt with in a documented white paper by this
report’s senior author.892 Capable technologists worldwide would not
already have created 172 prototype hydrogen cars and 87 hydrogen filling
stations893 if they were simply sloppy thinkers or deluded dreamers.
Rather, they have developed practical solutions to the problems that
many recent studies (often propagating each other’s errors) carelessly
assume to be formidable or insoluble. As energy venture capitalist Robert
Shaw says of the hydrogen transition, “Those who think it can’t be done
shouldn’t interrupt those doing it.”
892. Lovins 2003b.
893. Cars from
www.hydrogen.org/
index-e.html, filling stations
from M.P. Walsh 2004a.
230
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Option 4. Substituting hydrogen: Hydrogen: practical after all
Some of the business leaders now doing it were quoted in the frontispiece
of this book. A few additional quotations reinforce the impression that they
are doing it not for amusement but with the intention of making money:
“Our vision is that, from the year 2020, more than a third of all BMW vehicles
sold in Europe will be hydrogen-powered,” says company chairman Joachim
Milberg….Ferdinand Panik, director of DaimlerChrysler’s fuel-cell project in
Germany, reckons hydrogen fuel cells will power a quarter of new cars worldwide by 2020.
— ATLAS OF POPULATION & ENVIRONMENT,
AMERICAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE 894
“General Motors is on record saying that we expect to begin selling hydrogen
fuel cell vehicles by 2010, and we hope to be the first manufacturer to sell one
million fuel cell vehicles….[A]s the price of hydrogen and electricity per kilowatt
decrease, we see the convergence of the power and transportation infrastructures. The new model will cause hydrogen and electricity—made from various
sources—to become the two principal energy carriers. Hydrogen and electricity
are interchangeable via fuel cells and electrolyzers—leading to a distributed
energy network wherever one or the other is present—and also leading to
unprecedented new business opportunities.”
Capable technologists worldwide have
developed practical
solutions to the
problems that many
recent studies
carelessly assume
to be formidable
or insoluble.
“Those who think it
can’t be done
shouldn’t interrupt
those doing it.”
— LARRY BURNS, VP R&D AND PLANNING,
GENERAL MOTORS CORPORATION, 10–11 FEB. 2003895
“We are basically moving to a hydrogen economy.***[W]e believe the hydrogen
fuel cell is the big answer.”
— RICK WAGONER, CHAIRMAN AND CEO,
GENERAL MOTORS CORPORATION, 2003896
“The fuel cell is the most promising option for the future. We are determined to
be the first to bring it to market.”
— JÜRGEN HUBBERT, DAIMLERCHRYSLER897
“Work on the fuel cell is no longer motivated exclusively by technological and
environmental considerations, but has become a genuine competitive factor.
We view the fuel cell as an economic opportunity that will help safeguard hightech jobs and business success in the future.”
— DR. FERDINAND PANIK, FUEL CELL PROJECT DIRECTOR,
DAIMLERCHRYSLER, 12 NOV. 1999 898
“Fuel cell vehicles will probably overtake gasoline-powered cars in the next 20 to
30 years.”
— TAKEO FUKUI, MANAGING DIRECTOR, R&D,
HONDA MOTOR CO.899
“We want to meet our customers’ needs for energy, even if that means leaving
hydrocarbons behind.”
— SIR MARK MOODY-STUART,
CHAIRMAN, ROYAL DUTCH/SHELL GROUP 900
Billions of dollars have already been invested in hydrogen fuel cells by
most of the world’s major automotive and oil companies (GM, DaimlerChrysler, and Shell have each placed billion-dollar bets, with major efforts
also at Toyota, Honda, and BP), scores of leading manufacturing and
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
894. Pearce & Boesch,
undated.
895. Burns 2003.
896. First portion from
Sehgal 2003, second from
Wagoner 2003.
897. Hubbert, undated.
898. DaimlerChrysler
Communications 1999.
899. Fukui 1999.
900. Economist 2001.
231
Implementation
Option 4. Substituting hydrogen: Hydrogen: practical after all
materials companies (such as United Technologies and DuPont), and the
large existing hydrogen industry (Air Products & Chemicals, LindeAG,
Air Liquide, and more). Some of the most formidable potential competitors haven’t yet even announced their presence. Automakers alone are
continuing to invest more than a half-billion dollars a year to bring fuel
cells to market; they’re interested not just in fuel cells’ doubled efficiency
and zero emissions but also in their ruggedness, light weight, and simplicity. A fuel-cell powertrain has about a tenth the parts count of an enginebased powertrain, and with some 700 million cars and trucks in the world,
increasing about 8% a year, that’s a lot of avoidable parts. Then there is
substantial R&D by the military, which has long used fuel cells in submarines and spacecraft and is eager to apply them in land, sea, and air
platforms. Many of these investments have been to good effect, as will
become clear from announcements over the next year or so. Steady progress with the details of materials and manufacturing engineering is now
making fuel cells sufficiently durable (already over 10,000 hours in some
realistic field tests) to support commitments to the manufacturing scale-up
needed to make them cheap.
901. Solomon & Banerjee
2004.
The United States’ $1.2-billion hydrogen R&D commitment is following
in the footsteps of substantial hydrogen R&D efforts901 in Canada, Japan,
Europe (with special efforts in Germany, Norway, and Iceland), South
Korea, China, and elsewhere; California and Michigan even have significant state-level hydrogen programs. The efforts in China and Japan merit
special notice because of these countries’ competitive threat in automaking (pp. 132–136, 166–168). China is believed by some analysts to be
already the world’s number two hydrogen producer. Japan’s R&D investment, proportionate to the United States’, started earlier and is well integrated across fueling infrastructure, vehicles, and stationary uses. Its
aggressive and specific goals include 50,000 fuel-cell vehicles by 2010 and
5 million by 2020, plus 2.1 and 10 GW of stationary power generation by
those years (respectively including ~1.2 and ~5.7 million household-size
fuel cells). Model hydrogen-related building and safety regulations are to
have been redesigned by 2004, and Japan’s industrial standards for
hydrogen equipment are intended to shape global ones. These are among
the hallmarks of a serious strategic effort, and they are intensifying.
Fuel cells, in short, have already passed the stage of potential show-stoppers and are well along the path of serious commercialization. So, as we’ll
suggest, are the other parts of the hydrogen system needed to sustain a
practical route from here to there. So in broad outline, what would a
hydrogen transition look like? How could it enable the United States to
rely completely on domestic energy sources to fuel its economy not just
while natural gas resources last, but indefinitely? And from the perspective of broad national benefit, how could its radical improvements in
three sectors—vehicles, energy distribution, and overall energy infrastructure—make the hydrogen transition rapid and profitable?
232
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 4. Substituting hydrogen
Eight basic questions
Why is hydrogen important, and is it safe?
Hydrogen is important because it is a new, extremely versatile, and abundant energy carrier that is cleaner, safer, and potentially more economic
to use for mobility and distributed applications than either natural gas or
oil. As a carrier, hydrogen is made from other energy sources such as gas,
oil, coal, or electricity (including renewably produced electricity), and can
deliver any kind of energy services. The hydrogen industry is already
large—it makes about 50 million tonnes of hydrogen a year, or one-fifth
as many cubic feet as the world’s entire natural-gas production—yet has
an enviable safety record spanning more than a half-century. Any fuel is
hazardous and needs due care, but hydrogen’s hazards are different and
generally more tractable than those of hydrocarbon fuels.902
How would a light vehicle safely and affordably store enough
hydrogen to drive 300+ miles?
Inexplicably, recent studies from such normally informed sources as the
Office of Technology Policy, National Academy of Sciences, and American
Physical Society (not actually independent studies, but largely copied from
each other) have claimed that hydrogen storage is an unsolved problem,
a “potential showstopper,” and requires basic materials breakthroughs.903
They’re wrong; the storage problem has been solved for at least four years.
The solution, though, is not to find a new way to store hydrogen, but to
adopt vehicles whose better platform physics let them drive as far using
three times less hydrogen. For example, in 2000 the Revolution concept-car
design showed that properly integrating ultralighting, reduced aerodynamic drag and rolling resistance, and other improvements can triple vehicle efficiency without changing the fuel and prime mover. This reduction
in the energy needed to move the vehicle largely overcomes the roughly
fourfold904 higher bulk of compressed hydrogen gas (at the 5,000 psi that’s
now standard in off-the-shelf and extremely safe carbon-fiber tanks) vs.
gasoline for the same driving range. The skeptics simply forgot to count
the potential to make the vehicle more efficient, as distinguished from its
powertrain. But using an ultralight (optimally an advanced-composite)
autobody not only makes the hydrogen tanks small enough to fit or “package” well, leaving plenty of room for people and cargo, as shown in
Fig. 39; it also makes the fuel cell three times smaller. That makes it competitive many years earlier, because needing three times fewer kilowatts,
Implementation
Hydrogen’s rationale,
market, safety,
storage, cost, infrastructure, ultimate
energy sources, and
transitional path are
all solved or soluble
problems. The logical
sequence is from
today’s hybrid cars to
ultralight gasoline
hybrids to ultralight
hydrogen hybrids—
preferably using fuel
cells, but potentially
with internal-combustion engines as an
optional backup.
The storage problem
has been solved for
at least four years.
The solution, though,
is not to find a new
way to store hydrogen, but to adopt
vehicles whose better platform physics
let them drive as far
using three times less
hydrogen.
902. See NHA, undated; EERE, undated. It follows that the risks of wide public deployment are comparable to or less than those of the existing wide public
deployment of other fuels, including gasoline. However, historic doctrines governing tort liability may not adequately recognize this (Moy 2003). The United
States also has some 44,000 code jurisdictions, each with a fire marshal with probably a different and often an underinformed conception of hydrogen safety, so considerable effort will be needed, and is starting in the U.S. and abroad, to modernize codes and standards.
903. E.g., Service 2004. See Lovins & Williams 1999; Lovins & Cramer 2004; Lovins 2004b.
904. The compressed hydrogen actually has an eighth the energy per unit of volume of gasoline, but is used at least twice as efficiently in a fuel cell as
gasoline is in an engine.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
233
Implementation
Option 4. Substituting hydrogen: Eight basic questions: How to store enough hydrogen to drive 300+ miles?
Figure 39: Ultralight vehicles solve the hydrogen storage problem
Packaging for the fuel-cell Revolution concept-SUV virtual design described on
pp. 62–68 shows that such efficiency-tripling platform physics can shrink the
hydrogen tanks by threefold (the three tanks shown provide a 330-mile average
driving range on 3.4 kg or 138 L of hydrogen at 5,000 psi or 345 bar), yet still
offer a spacious interior (five adults in comfort, and up to 69 ft3 of cargo space
with the rear seats folded flat). The fuel cell (in rear) also becomes three times
smaller and more affordable. This design used three hydrogen tanks only for an
analytically convenient match to standard sizes; one or two tanks in custom
sizes would be lighter and cheaper, and conformal shapes are also possible.
Ford and others have demonstrated that such tanks are extremely crashworthy, partly because they’re supported by interior pressure. These governmentapproved tanks are standard in the industry; some makers are starting to use
10,000 psi, but ultralight vehicles make this higher pressure and cost unnecessary. The transverse tanks in this design have room to move axially in a sideimpact collision. The exterior of the Revolution show car is Fig. 18d on p. 61.
See also Box 7, pp. 62–63 above.
the automaker can afford to pay
three times as much per kilowatt—
a price realized at probably tens of
times lower cumulative production
volumes.905
Under what conditions is
hydrogen a cheaper
light-vehicle fuel than oil?
What matters is the total cost per
mile driven, not the cost of the fuel
per unit of energy contained. Since
hydrogen must be made from other
forms of energy, it is economic to
deploy only if it is more efficiently
used than the fuel it was made
from, thus justifying the investment, energy input, and other
operating costs needed to convert
it. Like gasoline made from crude
oil, or electricity made from coal,
hydrogen is a more “refined,” efficiently usable, valuable, and costly
form of energy than the natural
gas, electricity, or other resource it’s
made from. But if the price of
hydrogen per BTU is, say, twice
that of gasoline, but each BTU of
Source: Graphic from Lovins & Cramer 2004 (which describes this design), courtesy of Hypercar, Inc.
hydrogen also propels the car twice
as many miles as a BTU of gasoline, and if the car is priced the same, then the driver is indifferent. Or if
the car is priced higher but the hydrogen has a sufficiently greater efficienWhat matters is the
total cost per mile
cy edge, then the driver is again indifferent. Of course, car buyers are not
driven, not the cost of
so simple-minded as to care only about cost per mile: the enormous diverthe fuel per unit of
sity of vehicles on the market demonstrates how many other factors matenergy contained.
ter too (and often matter more). But just to focus for the moment on cost
per mile, Fig. 40 illustrates the relationship between fuel price, fuel-cell
905. At an 80% experience curve, meaning that each doubling of cumulative production volume reduces real unit cost by 20%, a two-thirds cost reduction is
achieved by five doublings of cumulative production volume. Five doublings are a factor of 32.
906. A good fuel cell is inherently far more efficient than a good gasoline engine and modestly more efficient than a good diesel engine. The fuel cell’s most
efficient operating point is also at the relatively low power at which most driving occurs, rather than at high power as for Otto engines (Williams, Moore, &
Lovins 1997, Fig. 8). Both engines and fuel cells can be advantageously hybridized to match their “map” (efficiency vs. load) with varying tractive loads and to
recover braking energy for re-use. Hybridization usually brings greater percentage savings to engines than to fuel cells, but this is sensitive to many assumptions about hardware, driving cycles, and control algorithms. The fuel-cell vehicle analyzed here combined a 35-kW fuel cell with a 35-kW high-power buffer
battery, mainly because for the time being, batteries cost less per kW than fuel cells. When that ceases to be true, the battery will become smaller. Ultimately
it may disappear if developers can perfect a reversible fuel cell, whose electrolytic (backwards) operation to store braking energy would be buffered with an
ultracapacitor for charge-acceptance rate. The byproduct oxygen could optionally be stored and used to supercharge the fuel cell, making it smaller.
234
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Option 4. Substituting hydrogen: Eight basic questions: Hydrogen a cheaper light-vehicle fuel than oil?
This comparison assumes that the
fuel-cell vehicle will sell for 23%
more than the gasoline-hybrid
version. Fig. 20 on p. 65 shows that
this premium is entirely due to
the fuel-cell powertrain’s costing
~$6,000 more to manufacture than
the gasoline-hybrid powertrain.
So long as that remains true, and the
retail price premium for the fuelcell vehicle remains $10,300, Fig. 40
shows that the fuel-cell version
can’t save enough fuel to repay its
higher capital cost. However, this
could, and probably will, change in
any or all of three ways by the time
the fuel-cell vehicles get to market
sometime during the following
decade:
Figure 40: How can fuel-cell vehicles compete with gasoline hybrids?
The toughest competitor will be not inefficient, heavy, gasoline-engine cars but
ultralight hybrids, assuming the prices and efficiencies of the gasoline-hybrid
and fuel-cell-hybrid midsize crossover-style SUV variants shown in Fig. 20, p.
65. Hydrogen delivered into such a fuel-cell ultralight SUV for $4/kg can compete against a gasoline-hybrid SUV with equal platform physics (mass, drag,
and rolling resistance) only if fuel cells become cheaper than we assumed, or
if, as shown in the lowest curve, they have durable membranes and the vehicles are plugged in as power stations when parked and earn appreciable revenue from Vehicle-to-Grid (V2G) operation. The range of V2G credits shown is
from Lipman et al.,907 and reflects a range of 10,000 to 40,000-h fuel-cell life and
less or more favorable spreads between natural-gas and electricity prices. The
high V2G revenues that yield the lowest cost curve in this graph could instead
be considered a surrogate for capturing some of the distributed benefits of
relieving congestion in the electric distribution grid.
0.55
total cost to drive ($/mile)
efficiency, and capital cost for the
two most efficient versions of the
Revolution ultralight midsize SUV
concept-car designs (Fig. 20 on
p. 65)—propelled by a gasolinehybrid-electric powertrain or a
hydrogen-fuel-cell hybrid906—
vs. their conventional 2004 Audi
Allroad 2.7T comparable competitor.
2004 Audi
Allroad
27T
0.50
0.45
gasoline
hybrid
0.40
hydrogen
fuel cell
no V2G
0.35
hydrogen
fuel cell
less V2G
0.30
0.25
0.20
0
0.50 1.00 1.50
2.00 2.50 3.00 3.50
4.00
hydrogen
fuel cell
no V2G
(best case)
gasoline price ($/gallon)
Source: RMI analysis.
• Fuel cells and other hydrogen equipment may become cheaper than
we assumed. One-third of the fuel-cell SUV’s extra retail price, nearly
enough to make the fuel-cell car compete with the hybrid, is due to
what now look like conservatisms in our assumed cost of the fuel-cell
system and hydrogen tanks.908 Most of the rest of the premium should
disappear at higher production volumes. At very high volumes when
fuel cells are fully mature, a fuel-cell car might ultimately even cost
less to build than a gasoline-engine car; that’s certainly the intention
of the world’s main automakers.
907. Lipman, Edwards, &
Kammen 2004.
908. The most detailed public-domain mass-production cost analysis (originally from Directed Technologies, Inc., which did much of Ford Motor Company’s
fuel analysis) states that this vehicle’s 35-kW fuel-cell system should cost about $1,723 at an early production volume of 300,000/y, and its tanks, to hold its
3.4 kg of hydrogen, should cost about $704 more than a normal gasoline tank: Ogden, Williams, & Larson 2004, p. 15, converted from 1998 to 2000 $. RMI’s
consultants’ cost analysis assumed a ~60% costlier fuel-cell system (excluding its output power electronics) and 150% costlier tanks than this, adding nearly
$4,000 to the assumed retail price shown in Fig. 20. (The fuel-cell system in this context means the stack plus auxiliaries—air compressors, heat exchangers,
humidification systems, safety devices, and control systems—but excludes the hydrogen tanks and associated plumbing.) Fig. 20 also shows a ~$600 lower
manufacturing cost for the fuel-cell than for the hybrid SUV in the category of chassis components, partly because the fuel-cell variant is 24 kg lighter than
the hybrid. We haven’t attempted to update the advanced-composites costs, which in light of recent manufacturing-process progress will probably be lower
than we assumed for all the variants.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
235
Implementation
Option 4. Substituting hydrogen: Eight basic questions: Hydrogen a cheaper light-vehicle fuel than oil?
Three ways this could
change (continued):
• Fuel cells may become more efficient. Our analysis assumed a peak
fuel-cell efficiency five percentage points lower than a 2003 norm (p.
62, note 328); this was a good economic choice, but further technological progress may make it unnecessary.909
Hydrogen fuel cells
today, installed in
the right places and
used in the right way,
can economically
displace less
efficient central
resources for delivering electricity, paving
the way for hydrogen
use to spread rapidly,
financed by its own
revenues.
• The vehicle may earn revenue from selling back power to the grid when
parked—so-called Vehicle-to-Grid or V2G operation, which RMI proposed in the early 1990s.
Our insights into the full economic value of distributed power suggest
that hydrogen fuel cells today, installed in the right places and used in the
right way, can economically displace less efficient central resources for
delivering electricity, paving the way for hydrogen use to spread rapidly,
financed by its own revenues.910 This logic normally refers to stationary
uses—combined power generation, heating, and perhaps cooling in buildings and industry. But it could also apply to power generation in parked
fuel-cell vehicles designed for this purpose. Using a more durable fuel
cell (whose extra cost we’ve counted), such plug-in “power stations on
wheels” could sell power to the grid when and where it’s most valuable—during afternoon peaks in downtown load centers. The first couple
of million adopters may even earn back as much as they paid to buy their
cars. The fuel cells in a full hydrogen-fuel-cell light-vehicle fleet would
ultimately have very many times the grid’s total generating capacity, so it
doesn’t take very many V2G-operated vehicles adopters to put most coal
and nuclear power plants out of business. Even at relatively high volumes, the power’s resale value to congested urban grids can be significant (Figs. 40–41).
What’s the cheapest way to produce and deliver hydrogen to
meet the economic conditions required for adoption?
We envisage a distributed model for hydrogen production and delivery
that integrates the infrastructures of natural gas, electricity, buildings, and
mobility. Instead of building a costly new distribution infrastructure for
hydrogen, or repeating the mistake of burdening supposedly cheaper central production with prohibitively costly distribution, we’d use the spare
offpeak capacity inherent in existing gas and electricity distribution infrastructures, then produce the hydrogen locally—chiefly at the filling station, called “forecourt” production—so it requires little or no further distribution. Hydrogen could thus be made, compressed, stored on-site, and
909. Many firms are developing materials that can run more efficiently by boosting operating temperature without unduly degrading membrane life. One major
developer states that a 5˚C-higher temperature can make a typical fuel-cell car 100–200 kg lighter—a big win if the membrane lasts.
910. Swisher 2003; Lovins et al. 2002. There are obvious applications in commercial buildings that need ultrareliable power for computers and other digital
loads, such as the well-known Omaha credit-card processing center, or buildings isolated from the grid, such as the Central Park police station (both successfully using ~$3,000/kW phosphoric-acid fuel cells). But there are also important early niche markets in industry. Microchip fabrication plants need ultrareliable
power, cooling, ultrapure hot water, and hydrogen—all valuable outputs of a fuel cell and its on-site natural-gas reformer, making even retrofits an attractive
proposition that several chipmakers are now considering. GM and Dow are also implementing a large pilot project that uses fuel cells to turn byproduct hydrogen from electrochemical processes back into electricity and useful heat to run the process. Such opportunities abound throughout industry.
236
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Option 4. Substituting hydrogen: Eight basic questions: Cheapest way to produce and deliver hydrogen?
delivered into vehicles for ~$4/kg
(less with cheap natural gas or at
Costco scale) with small-scale, onsite natural gas reformers. This
assumes that delivered gas prices
remain below $8/MBTU—a very
safe assumption in our view,
whether based on North American
gas or on LNG, which would nominally deliver gas to a filling station
for ~$6/MBTU. We assume conventional miniature thermal steam
reformers, which several firms are
now bringing to market. Their natural-gas-to-hydrogen conversion
efficiency is 72–75%. If a new type
of one-step reformer with lower
capital cost and higher efficiency
(83–85%) scales down to this size,
its production costs would be significantly lower than Fig. 42 shows.
Figure 41: Cost of distributed hydrogen production
Based on ~2003 technology for miniature reformers and electrolyzers, both
sited at the filling station, RMI analysis shows that the electricity or gas prices
shown can support delivery of 5,000-psi hydrogen into light vehicles at the
approximate $/kg costs shown for two different filling-station sizes, one standard (serving about as many cars as a normal million-gallon-a-year gasoline
filling station) and one hypermarket-scale (ten times as big). The dashed line
shows that a 500-vehicle urban parking facility, equipped with a 1-MW stationary fuel and a gas reformer for refueling, could deliver hydrogen at an intermediate cost (undercutting a standard filling station by ~$0.5/kg) if it captured the
revenues from an average of 10 kW of net power resold from each parked
vehicle, 10 h/d, 250 d/y.
cost of H2 ($/kg)
electrolysis
methane steam
reformer
9.00
ty p i
8.00
cal
h yp
serv
e rm
ic e
a rk e
7.00
s ta t
io n
t sc
V2G urban
fueling center
6.00
5.00
ale
4.00
3.00
2.00
1.00
0.07
0.06
0.05
0.04
0.03
power price ($/kWh retail)
4.00
5.00
GM estimates that 11,700 forecourt
Source: RMI analysis.
reformers could provide a station
within two miles of 70% of U.S. population, and permit refueling every
25 miles along the 130,000-mile National Highway System (NHS)—all for
a total investment of just $12 billion, assuming rather costly facilities.911
This would eliminate the “chicken-and-egg problem” (which comes first,
the fueling stations or the cars they fuel?) at less than a tenth the real cost
of the NHS itself. The 6,500 non-NHS stations in 100 metro areas could
fuel 1 million vehicles, equivalent to several million at the more fuel-frugal State of the Art platform efficiency. More stations would then be added
as needed to serve growing demand. (The U.S. now has ~170,000 gasoline
filling stations, and their number is shrinking.) The roll-out needn’t take
long: Deutsche Shell said a few years ago that if desired, it could put
hydrogen pumps into all its German filling stations in about two years.
Roughly 70% of gasoline filling stations, serving ~90% of U.S. gasoline
demand, already have natural gas service. The rest could reform their
hydrogen from either LPG or ethanol, which has the advantage that it can
be wet (hydrous)—avoiding the costly step of removing virtually all the
water—because the reformer needs steam as well as ethanol.912 As for climate impact, standard miniature reformers would emit ~2–3 times less
CO2 per mile that gasoline cars emit today. Several oil companies even
think they’ll find ways to collect that CO2 from filling-station-scale reformers and turn it into value. And of course biofuel reformers wouldn’t affect
the climate at all.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
6.00
7.00
8.00
9.00
gas price ($/Mbtu delivered)
Instead of building
a costly new distribution infrastructure
for hydrogen, or
repeating the mistake
of burdening supposedly cheaper central
production with
prohibitively costly
distribution, we’d
produce the hydrogen
locally—chiefly at
the filling station—
so it requires little or
no further distribution.
911. McCormick 2003.
912. Deluga et al. 2004;
Morrison 2004. Intelligent
Energy Inc., chaired by
former Shell chair Sir John
Jennings, has even demonstrated a shoebox-size direct
ethanol fuel cell (www.
intelligent-energy.com).
237
Implementation
Option 4. Substituting hydrogen: Eight basic questions: Cheapest way to produce and deliver hydrogen?
Alternatively, sites that can buy electricity for under 3¢/kWh—mostly during
off-peak periods or in areas with excess hydroelectricity or wind power—
may find on-site electrolysis competitive.913 These two ubiquitous and competitive retail commodities, natural gas and electricity, will compete in both
spot and forward markets, and the market will determine the winner. We
fully expect the hydrogen transition to be initially fueled by natural gas.
Ultimately, as renewables (chiefly windpower) become cheaper and natural
gas costlier, renewably produced electricity may gain the advantage.914
Fig. 41 shows a reasonable estimate for the cost competition between
distributed gas reformers and electrolyzers using near-term technology.
Are there enough primary energy sources for this transition?
After State of the Art efficiency improvements are deployed to the degree
practical by 2025, plus biosubstitutions, the U.S. will consume 20.4 Mbbl/d
of oil in 2025. Of this, light vehicles and trucks will consume 63% or 12.9
Mbbl/d (25.4 qBTU/y).
913. Electrolyzers are
somewhat like fuel cells run
backwards, so they too
should become much
cheaper with mass-production. They may have a cost
advantage at small scale,
e.g., for home use or very
small filling stations.
914. For niche markets and
perhaps more, one would
then need to consider
whether advanced lithium
batteries, which become
ever better and cheaper for
cellphones and laptop computers, might get cheap
and light enough so that an
advanced battery-electric
powertrain might beat a
fuel-cell one. See
Technical Annex, Ch. 24.
915. Heavy vehicles start
with good diesel engines,
which are more efficient
than gasoline engines, but
would probably be replaced
not with proton-exchangemembrane (PEM) fuel cells
but with more efficient
high-temperature (e.g.,
solid-oxide) fuel cells, perhaps with topping and bottoming cycles added.
Doubled efficiency is therefore still an approximately
realistic starting-point for
this estimate.
238
Is there enough natural gas in North America to fuel the hydrogen transition without resorting to LNG imports? Our previous analysis says yes,
both because so much gas can be profitably saved and because fuel-cell
vehicles are so efficient that they need relatively little gas to make their
hydrogen. If hydrogen fuel-cell road vehicles, both light and heavy, were
twice as efficient as gasoline- and diesel-fueled State of the Art technologies,915 they’d need in 2025 only 8 qBTU/y (or 70 million tonnes a year) of
hydrogen delivered to the tank. To reform this much hydrogen from natural gas at state-of-the-art 85% efficiency would require 9.4 qBTU/y (9.2
TCF/y) of natural gas. These numbers decline rapidly as more efficient
vehicles are phased in. We showed in Fig. 21 on p. 66 that an average State
of the Art light vehicle at EIA’s 2025 sales mix would get 73 mpg. If hydrogen fuel cells improve this gasoline-hybrid fuel economy proportionately
to the Revolution SUV variants shown in Fig. 20, p. 65, then that 73 mpg
would become 119 mpg-equivalent. However, since the fuel would switch
from gasoline to hydrogen, the fuel-cell vehicle would not just reduce but
eliminate the hybrids’ remaining oil use, which is 4.4 Mbbl/d for all light
vehicles in 2025 and 2.2 in 2035 (extrapolating EIA’s forecast of vehicles
and vehicle-miles from its end in 2025). Of course, as fuel-cell vehicles
phase in over time, the vehicle stock will also be getting more efficient—
as it must do for the hydrogen to be attractive. Thus if we could substitute
a fuel-cell vehicle for all light vehicles on the road in 2035 (which will virtually all be State of the Art ultralight hybrids as shown in Fig. 36i, p. 181),
then the entire road-vehicle stock would need only 5.8 qBTU/y (51 million
tonne/y) of hydrogen, which could be made from 6.6 TCF/y of natural gas.
As our discussion of natural gas showed on pp. 112–116 above, improved
electric and direct-gas end-use efficiency and electric load management
could cost-effectively save 12.0 TCF/y of gas by 2025—a value we consider conservatively low. Further, since we are no longer making gasoline in
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 4. Substituting hydrogen: Eight basic questions: Are there enough primary energy sources for this transition?
Implementation
Figure 42: Reducing EIA’s 2025 natural-gas demand forecast by 41% via electric and gas efficiency, substitution of fuel-cell combinedheat-and-power (“FC CHP”) to back out the remaining combined-cycle gas power stations, gas savings in refineries (displaced by
substituting hydrogen for hydrocarbon fuels), and displacement of gas in petrochemicals by biomaterials (or offshore migration).
Naturally, unless directed elsewhere by government intervention, saved gas will flow to the uses with the highest netback margins.
Competition between the cost of extracting or saving gas and the value of saving it (then reusing it for cogeneration, hydrogen production,
process heat displacements, etc.) will determine whether all the forecast gas production occurs.
trillion cubic feet per year (TCF/y)
35
31.4
30
25
20
23.5
4.8
0.5
2.7
8.1
2025 N.A. production
0.2
3.5
2025 domestic production
3.9
2.5
10 TCF/y can be used
either to substitute
for oil or power
the H2 transition
15
10
19.5
23.9
19.2
16.7
domestic
Canada
LNG
5
savings
remaining
demand
refinery &
petrochemical
savings
remaining
demand
gas
efficiency
fuel cell CHP
power
efficiency
2025 gas
demand
2004 gas
demand
0
Source: EIA 2003, 2004; RMI analysis.
a hydrogen world, natural gas that is currently used to make hydrogen in
oil refineries could be redeployed, saving 0.5 TCF/y of gas. In addition,
we would expect at least two-thirds of U.S. refineries to close (since there
is no longer much market for petroleum-based transportation fuel, and
the remaining refineries would supply the remaining petroleum-fueled
vehicles, airplanes, and machinery), saving an additional 1.0 TCF/y of natural gas from process heat and power. This would increase further if aircraft shifted to liquid hydrogen—a surprisingly practical and attractive
option.916 One additional TCF/y would be saved by the eventual offshore
migration of U.S. petrochemicals manufacturing, or—as we would prefer—through substitution by biomaterials. This brings 2025 gas demand
down from 31.4 TCF to only 18.5 TCF. That’s 8.2 TCF below the forecast
2025 North American gas deliverability of 26.6 TCF (Fig. 42). Thus we
A sensible hydrogen
transition may even
require less financial
capital than
business-as-usual.
916. Liquid hydrogen is the lightest known fuel, four times bulkier but 2.8 times lighter than jet fuel per unit energy, or about as dense as high-density
Styrofoam—hence its use in space rockets. Using it to fuel jetliners requires superinsulated tanks, which are bulky but light, permitting ~20–25% higher payloads. Design studies for such “cryoplanes” by Boeing, NASA, Airbus (for the EU), and others confirm cleaner burning and better safety than jet fuel
(kerosene). Although Airbus thought efficiency would drop, Boeing (for a nominal 767 platform) found a ~10–15% gain—enough to offset the liquefaction
energy—because of hydrogen’s net effects on mass, drag, climb, cruise, and engine efficiency: Daggett 2003, and personal communication, 16 June 2003.
The Air Force tested liquid hydrogen in a B-57 in 1956; Tupolev, in a Tu-154 in 1988. Airbus’s 35-partner consortium (dieBrennstoffzelle.de, undated), under EU
funding, has already established the concept’s basic feasibility and safety. Boeing has announced work on fuel-cell applications for both propulsion and auxiliary power, and expected in 2003 to flight-test an experimental one-seat aircraft with propellers driven by a 25-kW fuel cell after battery-boosted takeoff
(Knight 2003). Dutch researcher Ing. P. Peeters believes the ultimate solution, especially for regional and other smaller airplanes, would be a fuel-cell, superconducting-electric-motor, unducted-fan-driven design (personal communication, 3 May 2004). At Mach 0.65, he predicts energy savings of 55% vs. 747-400
(7,000-km mission, 75% load factor, 415 seats) and 68% vs. 737-400 (1,000 km, 70% load factor, 145 seats). This implies substantial savings, vs. even a very
efficient future kerosene-turbofan design.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
239
Implementation
Option 4. Substituting hydrogen: Eight basic questions: Are there enough primary energy sources for this transition?
have enough gas supply to fuel the hydrogen transition in 2035, even
before all the State of the Art light vehicles are phased in. The gas savings
are summarized in Fig. 42.
The Dakotas alone
could make 50 million
tonnes of hydrogen
per year—enough,
at State of the Art
efficiency, to fuel
every highway
vehicle in the
United States.
This analysis assumes that all of the needed hydrogen would be reformed
from natural gas. In fact, some hydrogen would be made from electrolysis,
particularly using off-peak power, low-cost hydropower (such as spring
spillpower), and renewables. In fact, the wind potential of the Dakotas alone
could make 50 million tonnes of hydrogen per year917—enough, at State of the
Art efficiency, to fuel every highway vehicle in the United States.918 Dakotas
windpower is only a fraction of the national potential: adding Texas, Kansas,
and Montana augments the Dakotas’ output by 1.5-fold. Windpower is now
a $7-billion-a-year global industry. It adds 2% a year to Germany’s total supply of electricity (now 10% renewable), provides a fifth of Denmark’s, could
provide upwards of half Europe’s residential electricity by 2020, and now
includes such giants as General Electric. Obviously windpower can make
enough all-American hydrogen, plausibly at competitive long-run cost,919
to do everything we’re proposing here to do with natural gas. This makes
frontier gas projects (Rockies wildlands, North Slope, etc.) optional. Nearly
unlimited amounts of hydrogen could also be produced from coal, which is
very good at pulling hydrogen out of steam, if emerging ways to keep the
carbon permanently out of the air fulfill their promise. And finally, although
it’s too early to tell for sure, there are tantalizing indications that a sensible
hydrogen transition may even require less financial capital than business-asusual, partly because it needs less fuel and partly because upstream investments tend to be lower for gas than for oil.
917. Elliott et al. (1991) estimated the Dakotas’ Class 3+ wind potential, net of environmental and land-use exclusions (50%
of forest area, 30% of agricultural and 10% of range lands, 20% of mixed agricultural/range lands, 10% of barren lands, and
100% of urban, wetlands, and parks and wilderness areas), at 2,240 TWh/y (equivalent to 58% of total U.S. 2002 net generation) at 50-m hub height for 750-kW turbines. Today’s ~2-MW turbines have 100-m hub heights, where the wind is much
stronger (its extractable power rises as the cube of windspeed), and would normally do much better than the assumed
25% efficiency and 25% losses. Moreover, recently discovered larger-than-expected high-level wind would probably further increase the potential (Archer & Jacobson 2003). At a nominal 75% electrolyzer efficiency, the total wind electricity
from these two states could produce 50 million tonnes of hydrogen per year (Lower Heating Value = 120 MJ/kg) on-site,
less irrecoverable losses (perhaps on the order of 10%) in transmitting electricity or hydrogen to market. The electrolyzers
might also be able to sell byproduct oxygen for gasifying coal or biomass. At a nominal ~40% capacity factor, characteristic
of good but not outstanding wind sites with modern turbines, the wind capacity required would be ~640 GW, approaching
the total U.S. generating capacity of ~750 GW. This would be a considerable undertaking, but plausibly economic (Leighty
2003). When considering the expense, recall that the U.S. Senate has little trouble voting as much as tens of billions of dollars in subsidies for a clearly uneconomic pipeline to transport 35 TCF/y of stranded gas from Alaska’s North Slope. (Such a
pipeline, especially via the Canadian route, might make more economic sense if it carried hydrogen instead, reformed at
the wellhead with CO2 reinjection.) For general background on large-scale windpower, see www.awea.org;
www.ewea.org; Reeves 2003; Chapman & Wiese 1998; and L. Brown 2004.
918. In 2000, all U.S. highway vehicles used 20.7 qBTU of gasoline (77%) and diesel fuel (23%), 74% of it gasoline in light
vehicles (ORNL 2002, p. 2-6). With the more than quintupled-efficiency SOA-class fuel-cell light vehicles (119 mpg with fuel
cells vs. EIA’s 22.6) and roughly tripled-efficiency SOA-class fuel-cell heavy vehicles, that 22 EJ (exajoules = 1018 J) of
highway-vehicle petroleum fuel could be displaced by ~4.6 qBTU/y of H2, or 40 MT/y, leaving room for traffic growth.
919. Wind machines dedicated to powering electrolyzers could eliminate the cost, maintenance, and uptower weight of the
gearbox and power electronics.
240
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Option 4. Substituting hydrogen: Eight basic questions
What technologies are required to enable the hydrogen
transition?
The hydrogen transition depends on superefficient vehicles and distributed
generation taking hold in the U.S., more than either of these breakthroughs
depend on the hydrogen economy. There has been much misplaced angst
about whether the U.S. should invest now in efficient vehicles or in hydrogen technologies. This debate makes as much sense as arguing about
whether star athletes should play football or baseball, which occur during
different seasons. The answer is, of course, “both”: first today’s gasoline
hybrids, then ultralight hybrids, then ultimately fuel-cell ultralight hybrids.
(Some experts believe an intermediate step—efficient hybrids with small
hydrogen-fueled internal-combustion engines, like hybrid successors to
Ford’s Model U concept car—may also make sense; 920 these could be considered a partial backstop technology in case cheap, durable fuel cells take
longer to commercialize than expected.) In the case of hydrogen, efficiency
and distributed generation should clearly come first because these set the
stage for the hydrogen economy, which will have trouble competing without them. That is, hydrogen needs State of the Art-class vehicles far more than
they need hydrogen. However, once we have such vehicles and once fuel cells
become cheaper, there will be a robust business case for producing the
hydrogen that those vehicles would then use.
How can the U.S. profitably make the transition from oil
to hydrogen?
The transition will certainly be profitable for automotive and fuel cell
manufacturers. We have suggested elsewhere an orderly and integrated
sequence of deployment steps that can make it self-financing.921 However,
the hydrogen economy presents a paradox for oil companies. They will
already have lower total net income from efficiency adoption (if they sell
less oil but don’t invest in the efficiency improvements), although the rest
of society will have greater income and earnings. If they redefined themselves as energy companies, however, they could make more profit by
embracing hydrogen than by clinging to petroleum’s steadily declining
energy market share. Fundamentally, the margins on hydrogen could
be greater than the remaining margins on gasoline, for three reasons.
First, hydrogen will monetize gas reserves earlier, increasing their present
value. On a global basis, oil is becoming increasingly expensive to find
and exploit, while gas is comparatively plentiful but is often stranded
and remote from market. Second, fuel margins from hydrogen depend on
the efficiency of converting the fuel into torque at the wheels. The more
efficient the fuel cell is, the more suppliers can charge for the hydrogen,
because it competes at the wheels of the car, not per BTU but per mile.
Third, refinery assets will be underutilized due to decreased gasoline
production. Some of them are strategically located near urban centers
and therefore have the potential to become hydrogen production centers.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implementation
Hydrogen needs
State of the Art-class
vehicles far more
than they need
hydrogen. However,
once we have such
vehicles and once
fuel cells become
cheaper, there will
be a robust business
case for producing
the hydrogen that
those vehicles would
then use.
The margins on
hydrogen could be
greater than the
remaining margins
on gasoline.
Oil companies will be
able to make more
profit from their oil by
taking hydrogen out
of it in a reformer
than having to add
more hydrogen to it
in a refinery.
920. Thomas 2004.
921. Lovins & Williams 1999;
Lovins 2003b.
241
Implementation
Option 4. Substituting hydrogen: Eight basic questions: How to profitably transition from oil to hydrogen?
(Some refineries make more profit today as merchant electricity generators
than from selling refined products. Merchant hydrogen is an analogous
play and should be especially attractive for near-urban refineries because
they’ve already bought their methane steam reformers and other infrastructure; they would need only to deliver the hydrogen to nearby buyers.)
922. U.S. refineries use
~7 MT/y of hydrogen to
make high-octane gasoline
and to desulfurize diesel
fuel. These applications
worldwide are responsible
for most of the 11%/y growth
in hydrogen production.
The oil companies that position themselves to become future hydrogen
suppliers will survive the transition and prosper. Depending on the cost
of oil reforming technology and the relative cost of carbon credits vs. carbon sequestration, it may make more economic sense to reform rather than
refine oil: H minus C will be worth more than H plus C, or symbolically,
(H–C)>(H+C). Oil companies will be able to make more profit from their
oil by taking hydrogen out of it (and using it to split off more hydrogen
from steam) in a reformer than having to add more hydrogen to it in a
refinery.922 They’re already good at both these processes, but their emphasis would shift.
When could this transition occur?
Two keys will unlock
hydrogen’s potential:
early deployment of
superefficient vehicles,
which shrink the
fuel cells so they’re
affordable and the fuel
tanks so they package,
and integration of
deployment in vehicular and in stationary
uses, so each accelerates the other by
building volume and
cutting cost.
The hydrogen option
is not essential to
displacing most or
all of the oil that the
United States uses.
But it’s the most
obvious and probably
the most profitable
way to do this while
simultaneously
achieving other
strategic advantages.
242
The oft-described technical obstacles to a hydrogen economy—storage,
safety, cost of the hydrogen, and its distribution infrastructure—have
already been sufficiently resolved to support rapid deployment starting
now in distributed power production, and could be launched in vehicles
upon widespread adoption of superefficient vehicles. (The stationary fuelcell markets will meanwhile have cranked up production to achieve serious cost reductions, even if they capture only a small market share: twothirds of all U.S. electricity is used in buildings, and many of them present favorable conditions for early adoption.) Automotive use of fuel cells
can flourish many years sooner if automakers adopt recent advances in
crashworthy, cost-competitive, ultralight autobodies. We certainly believe
that the transition could be well underway by 2025, and if aggressively
pursued, it could happen substantially sooner. Two keys will unlock
hydrogen’s potential: early deployment of superefficient vehicles, which
shrink the fuel cells so they’re affordable and the fuel tanks so they package, and integration of deployment in vehicular and in stationary uses, so
each accelerates the other by building volume and cutting cost.923
In sum, the hydrogen option is not essential to displacing most or all of
the oil that the United States uses. But it’s the most obvious and probably
the most profitable way to do this while simultaneously achieving other
strategic advantages—complete primary energy flexibility, climate protection, electricity decentralization, vehicles-as-power-plants versatility, faster
adoption of renewables, and of course deeper transformation of automaking and related industries so they can compete in a global marketplace
that’s headed rapidly in this direction.
923. This thesis (Lovins & Williams 1999) is now integral to many major players’ business strategies, and is briefly summarized
in Lovins 2003b.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implications and Conclusions
ndgames in chess have just two players. Their dance of their
moves and countermoves shapes the outcome with an intricacy
that only the world’s finest grandmasters and fastest computers
can hope to anticipate, and that only in broad strokes. The Oil Endgame
has innumerable players, and its complexity is far too great to grasp or
foresee. But its implications merit discussion relating to employment,
allies and trading partners, developing countries, oil-exporting countries, oil companies, other energy industries, military affairs, the federal
budget, and environment and public health. We then conclude with
some broad lessons and next steps.
E
The opportunities
just described have
profound and encouraging implications
for virtually all stakeholders, turning many
old problems into new
win-win opportunities.
Implications
Investing in efficiency
and renewables tends
to make far more new
jobs directly than it
displaces old ones;
lower-cost energy
services will induce
respending, hence
further job creation
throughout the
economy.
Employment
A welcome effect of displacing oil with cheaper alternatives is significant
creation of good jobs, both to produce the biofuels and the oil-saving hardware, and through respending the saved oil expenditures, turning a cash
outflow into a domestic multiplier of jobs and income. Such increased
employment on a broad geographic and skill base is a vital sociopolitical
need in most industrialized countries. Renewable energy is already making new jobs in Europe, which expects 0.9 million net by 2020 (over half
from biofuels)924 and may get even more as the pace of switching to renewables accelerates (the current EU target for electricity is 20%-renewable by
2010). Denmark reportedly has about three times as many jobs from manufacturing wind turbines, in which it’s the world market leader, as from
its electric utility industry.
In the United States, our rough estimates suggest that by 2025, the highervalue vehicles being made by the revitalized automotive industry could
generate ~240,000 new automaking and supplier jobs;925 making their lightweight materials could generate possibly some more, due to higher valueadded (pp. 161–162); and producing biofuels, nearly 780,000 jobs just from
10% displacement of gasoline by ethanol by 2020, or nearly 1.5 million jobs
at our projected 2025 ethanol volume (p. 165). Inevitably, some jobs will also
be displaced. By 2020, ~86,000 jobs, mainly within the petroleum industry,
will be lost, with some positions such as petroleum engineers, refinery technicians, and pump operators eliminated.926 However, many of those skilled
924. ECOTEC 1999.
925. Estimated from EIA’s 21.6-million-light vehicle sales projection for 2025, 77% new-sales share of SOA vehicles in 2025
(Fig. 36i), 75% domestic manufacturing share (as in 2003 per Automotive News 2003, pp. 16, 20, 25), additional manufacturing
cost of $2,544 per unit (p. 70 above, note 345), the 1998 coefficient of 7.57 direct and supplier jobs (excluding respending multiplier) per million 2000 $ of automaking revenue (McAlinden, Hill, & Swiecki 2003, pp. 8, 16), and a ~0.5% sales loss from
gross price elasticity. These inputs yield 239,000 marginal automaking and supplier jobs due to the higher value-added to
make State of the Art vehicles. We didn’t try to calculate potential shifts, in either direction, in job intensity; this is just a firstorder approximation.
We estimate a net
increase in U.S. jobs,
due to efficiency
improvements, of more
than a million jobs
by the year 2025 just
from producing oilsaving hardware and
biofuels. Respending
the net saving in oil
dollars—$133 billion
a year by 2025, equivalent to a very large
tax cut—should also
stimulate employment
considerably.
926. Bezdek & Wendling 2003.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
243
Implications
Any country,
including
America’s even more
oil-dependent partners
abroad, can advantageously apply similar
oil-displacing opportunities, differing only
in detail. A strong U.S.
example will be
critical; expecting
others to do what we
say, not what we do,
would be hypocritical.
A helping hand, technical and policy
collaboration free of
“not-invented-here”
constraints, a friendly
rivalry in who can
save oil fastest,
and the spillover of
U.S.-led technology
acceleration into
global markets would
do wonders for rapidly
reaching a “tipping
point” that turns global
oil consumption
irrevocably
downward.
Employment
and versatile people are likely to be rehired by successor industries, such as
biorefineries, in the revitalized rural economy (pp. 162–165). Over all enduses (those just counted explicitly plus a conservative estimate for trucks,
airplanes, buildings, and industry), we estimate a net increase in U.S. jobs,
due to efficiency improvements, of more than a million jobs by the year
2025 just from producing oil-saving hardware and biofuels. Respending
the net saving in oil dollars—$133 billion a year by 2025, equivalent to a
very large tax cut—should also stimulate employment considerably.
Allies and trading partners
“Under current circumstances,” former CIA Director R. James Woolsey
recently testified, “an oil crisis will affect all our economies, regardless of
the source of our own imports. We must think in terms of the world’s
dependence, not only our own.”927 Moral, cultural, and historical ties
aside, there are compelling commercial reasons for encouraging allies and
trading partners to engage in their own versions of the Mobilization strategy. The U.S. invests heavily in the military security of Western Europe,
Japan, South Korea, Taiwan, and Israel, none of which has oil; indeed,
they depend on oil and oil imports even more than we do. Those relationships are important for other reasons. Many countries are more worried
about oil than the United States seems to be, and may well choose to
embark on a new path of aggressive oil displacement once they realize it’s
possible. The more oil such friends join in saving, the more the common
oil problem diminishes. America should therefore invest the persuasive
power of its own example, the creativity of its scientists and technologists,
the competitive skills of its businesspeople, and the energy of its citizens
in making the transition beyond oil a compelling global trend. The alternative is spending blood and treasure to get and keep access to oil that
others, for their own deeply held reasons, will increasingly seek to deny
to developed countries in general and to the United States in particular.
In our professional careers, the coauthors of this report have worked on
energy issues in the private and public sectors in more than fifty countries
worldwide. Each society has a unique economic structure, culture, and
climate, and they differ widely in styles of governance (e.g., laissez-faire
vs. dirigiste). Yet so far we have not found one to which a variant of the
technological and policy approach described here could not be effectively
adapted if skillfully tailored to local conditions.
The main obstacle to hatching home-grown versions in some countries has
been their erroneous conviction, born of their striking progress in saving
energy two or three decades ago, that they are already as energy-efficient
as they can be. Of no country on earth is that true. There are, of course,
927. Woolsey 2004.
244
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Implications
Allies and trading partners
many differences; savings in places like Western Europe and Japan may be
modestly smaller and costlier than in the U.S. But savings will still be large
and lucrative, because the differences are less important than the similarities. A strong U.S. example, a helping hand, technical and policy collaboration free of “not-invented-here” constraints, a friendly rivalry in who can
save oil fastest, and the spillover of U.S.-led technology acceleration into
global markets would do wonders for rapidly reaching a “tipping point”928
that turns global oil consumption irrevocably downward.
Developing countries
The World Bank estimates that 2.3 billion people today have no access to
electricity and 1.6 billion have no access to modern fuels. For many of
those people, life is nearly as nasty, brutish, and short as it was a thousand years ago. Such deprivation, hunger, preventable disease, and illiteracy are blots on the world’s conscience. Turning these conditions into
opportunities for shared wealth creation is an unimaginably large opportunity for which the pioneers of “Bottom of the Pyramid” thinking and
action make a powerful business case.929
Leapfrog development
Among the several billion people who are starting to get ahead, most
notably in the burgeoning commercial centers of China and India, the oildisplacing potential for leapfrogging over obsolete development patterns
is stupendous. Societies that are building housing, offices, factories, and
infrastructure and are producing appliances, vehicles, and industrial
equipment have the chance to do it right the first time and to adopt
world-class resource efficiency. In fact, failure to do so is one of the heaviest drags on development, because inefficient resource use diverts most of
the investment into costly supply-side projects, leaving too little to buy
the things that were supposed to use the resources. For example, the
financial capital needed to build a factory making quintupled-efficiency
compact-fluorescent lamps or heat-blocking superwindows is about a
thousandfold less, and pays back about ten times faster, than the capital
otherwise needed to expand the supply of electricity to provide the same
increase in light or comfort. The product of intensity times velocity of
capital implies about a ten-thousandfold saving from buying “negawatts”
wherever they’re cheaper than megawatts. This strategy—based on
rewarding electricity providers for reducing bills rather than selling energy—could turn the power sector, which now gobbles about one-fourth of
the world’s development capital, into a net exporter of capital for other
development needs.930
This shift is even
more vital and timely
for developing countries, which are over
twice as oil-intensive
as rich countries
but can afford oil
even less. Making
advanced energy
efficiency the cornerstone of the development process can
also free up enormous amounts of
scarce capital to fund
other development
needs. This strategy
may especially commend itself to China
and other Asian
nations eager to
escape or avoid the
oil trap.
928. This characteristic of nonlinear systems means that a process of change reaches the point where a small, seemingly
unimportant, incremental change tips the system into a wholly different mode of behavior. Malcolm Gladwell (2000) illustrates with a rhyme: “Tomato ketchup / in a bottle, first none’ll come, / and then the lot’ll.”
929. Prahalad 2004.
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
930. Lovins & Gadgil 1991
245
Implications
Developing countries: Leapfrog development
What does this have to do with oil? A lot. Building an oil-frugal or even
an oil-free economy from scratch is easier than converting an oil economy
to kick the habit. Superefficient energy use, built in the first time, actually
reduces the capital cost of many buildings and industrial processes; nonoil supplies suffer less hard-currency outflow and price volatility; and the
high capital intensity of oil-related supply investments is avoided. It’s
gratifying that many developing countries are eager to progress past the
oil trap without falling in, and disappointing that bilateral and multilateral aid and advice efforts are doing so little to help them execute this
emerging strategy. The opportunity is unprecedented; the prize is vast;
the time is short. China’s mid-2004 energy strategy, making efficient use
the top national priority (pp. 135–136, above) is an act of farsighted leadership that we hope others, in both the developing and developed
worlds, will emulate.
Climate change and development
“Developing countries
have the potential
to leapfrog the developed world’s process
of industrialization,
thereby providing an
enormous opportunity
to improve energy
efficiency and reduce
emissions.”
— Lord John Browne,
Chairman, BP plc
There’s a striking convergence between the goals of accelerating equitable
global development, averting oil dependence by those not already suffering from it, and helping developing countries avoid becoming as much of
the climate problem as developed countries now are. As Lord Browne,
Group Chief Executive of BP plc, recently wrote:931
It would be morally wrong and politically futile to expect countries struggling to
achieve basic levels of development to abandon their aspirations to grow and to
improve their people’s living standards. But it would be equally wrong to ignore
the fact that by 2025, energy-related carbon dioxide emissions from development
countries are likely to exceed those from the member states of the Organization
[for] Economic Cooperation and Development. Instead of being daunted by the
scale of this challenge, policymakers must recognize the scale of the opportunity:
developing countries have the potential to leapfrog the developed world’s
process of industrialization, thereby providing an enormous opportunity to
improve energy efficiency and reduce emissions.
He elaborated this theme in the same spirit as our case for displacing oil:
Seven years after the Kyoto meeting, it is becoming clear that the reduction of
greenhouse gas emissions is a soluble problem, and that the mechanisms for
delivering the solutions are within reach. In that spirit of cautious optimism,
it is time to move beyond the current Kyoto debate…. [T]he costs of deep-water
oil and gas development have fallen by a factor of three over the last 15 years,
dramatically extending the frontier of commercial activity. There is no reason to
think that research and development in the area of benign energy systems
would be less successful….Counterintuitively, BP found that it was able to reach
its initial target of reducing emissions by 10 percent below its 1990 levels without cost. Indeed, the company added around $650 million of shareholder value,
because the bulk of the reductions came from the elimination of leaks and waste.
Other firms—such as electricity generator Entergy, car manufacturer Toyota,
and mining giant Rio Tinto—are having similar experience. The overwhelming
message from these experiments is that efficiency can both pay dividends and
931. Browne 2004.
246
Winning the Oil Endgame: Innovation for Profits, Jobs, and Security
Developing countries: Climate change and development
Implications
reduce emissions.932…Neither prescriptive regulations nor fiscal interventions
designed to collect revenue rather than to alter behavior provide the answer.
Rather, governments must identify meaningful objectives and encourage the
business sector to attain them by using its knowledge of technology, markets,
and consumer preferences.
The global economy, oil savings, and development
We’ve already noted the salutary effects of biosubstitution on rural
economies and hence on the prospects for replacing agricultural subsidies, especially in the U.S. and EU, with real revenues for new rural outputs. This could be a big step toward letting developing-country farmers
compete fairly to serve both home and export markets (p. 210). The future
we envisage may or may not entail large-scale international trading of
biofuels, such as Brazilian ethanol. It would certainly involve dwindling
trade in oil—a roughly $400-billion business in 2002. But the decline of oil
as an important trade commodity should be more than offset by the rise
in trade for oil-displacing technologies and materials. Correcting highly
suboptimal investments in using oil and spending its proceeds should
also markedly improve global economic efficiency, vitality, and equity.
What would happen if not just other OECD nati